r-help-archive

r-help-archive - 26 new messages in 12 topics - digest

2009-12-17T13:16:00.001-08:00

R-help-archive
http://groups.google.com/group/r-help-archive?hl=en

Today's topics:

==============================================================================
TOPIC: cross-product
http://groups.google.com/group/r-help-archive/t/853af62ce8a99bf4?hl=en
==============================================================================

== 1 of 1 ==
Date: Thurs, Dec 17 2009 7:59 am
From: Marc Schwartz

On Dec 17, 2009, at 9:47 AM, Chuck White wrote:

> Hello -- I am trying to create a dataframe whose rows are the cross
> product of rows in two other dataframes. Here's an example:
>
> A = data.frame(F1=c('1','2','3','2'))
> B = data.frame(F2=c('4','5'))
> C = data.frame()
> for (x in unique(A[,'F1'])) {
> C = rbind(C, cbind(F1=x,B))
> }
>
> How can I achieve this without the for loop? Thanks.

See ?merge

> merge(unique(A), B)
F1 F2
1 1 4
2 2 4
3 3 4
4 1 5
5 2 5
6 3 5

BTW, avoid using 'C' as the name of an object, as C() is a function in
R. R will typically know the difference, but it is a good defensive
programming habit to avoid assigning objects to pre-existing function
names.

HTH,

Marc Schwartz

______________________________________________
R-help@r-project.org mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.

==============================================================================
TOPIC: StructTS standard errors
http://groups.google.com/group/r-help-archive/t/6805e80750fc9e16?hl=en
==============================================================================

== 1 of 1 ==
Date: Thurs, Dec 17 2009 8:13 am
From: Giovanni Petris

Hello,

Does anybody know if (and how) it is possible to obtain standard errors of estimated variances from StructTS? (R 2.10.0).

Thank you in advance,
Giovanni

==============================================================================
TOPIC: Error for making packages under windows XP-Error in library.dynam(lib,
package, package.lib)
http://groups.google.com/group/r-help-archive/t/44f9d82095e23cf3?hl=en
==============================================================================

== 1 of 1 ==
Date: Thurs, Dec 17 2009 8:13 am
From: "rusers.sh"

Dear Duncan,
I just followed an example to try making a package. The problem is really
on the useDynLib. It works after i delete it.
Thanks a lot.

2009/12/17 Duncan Murdoch <murdoch@stats.uwo.ca>

> rusers.sh wrote:
>
>> Hi,
>> I have installed the necessary tools for making a R package under windows
>> and am sure these tool have been correctly configured. I am very new to
>> try
>> writing a package, so the error may be very obvious for you. Forgive me if
>> it is too easy.
>> My package name is *stam*, and i put it under the folder
>> "D:/StatSoft/R/MyPackage/". I have two folders,*R* and *man*. Besides, i
>> also have *description* and *namespace** *files. It is a very simple beta
>> package.
>>
>>
>
> The DESCRIPTION and NAMESPACE files should be specified in upper case, but
> that's probably not your problem. I would guess you have a command that
> says to load a shared library, e.g. a useDynLib statement in your NAMESPACE,
> or a library.dynam in a .First() function. Just delete it, without compiled
> code (which would be kept in the src directory) you don't need it.
>
> Duncan Murdoch
>
> Following is the error after using "R CMD check mypackage". Thanks very
>> much.
>> #########
>> * using log directory 'D:/StatSoft/R/MyPackage/stam.Rcheck'
>> * using R version 2.10.0 (2009-10-26)
>> * using session charset: ISO8859-1
>> * checking for file 'stam/DESCRIPTION' ... OK
>> * this is package 'stam' version '1.0-0'
>> * checking package name space information ... OK
>> * checking package dependencies ... OK
>> * checking if this is a source package ... OK
>> * checking for executable files ... OK
>> * checking whether package 'stam' can be installed ... OK
>> * checking package directory ... OK
>> * checking for portable file names ... OK
>> * checking DESCRIPTION meta-information ... OK
>> * checking top-level files ... OK
>> * checking index information ... OK
>> * checking package subdirectories ... OK
>> * checking R files for non-ASCII characters ... OK
>> * checking R files for syntax errors ... OK
>> * checking whether the package can be loaded ... ERROR
>> Error in library.dynam(lib, package, package.lib) :
>> shared library 'stam' not found
>> Error: package/namespace load failed for 'stam'
>> Execution halted
>>
>> It looks like this package has a loading problem: see the messages for
>> details.
>>
>> #########
>>
>>
>>
>
>

--
-----------------
Jane Chang
Queen's

[[alternative HTML version deleted]]

==============================================================================
TOPIC: how to import data from excel to R
http://groups.google.com/group/r-help-archive/t/e2006930d493a3ed?hl=en
==============================================================================

== 1 of 8 ==
Date: Thurs, Dec 17 2009 9:24 am
From: sta_2279@spectrum.net.in

Hi,
I am using R and I want to know how data can be transferred from Excel
Spread sheet to R for analyzing. I have done like this

mydata<-read.table("C:\Documents and Settings\admin\Desktop\data.txt");

but its not working how can i do it

regards

Sarath Sankar V
Spectrum Softtech Solution

== 2 of 8 ==
Date: Thurs, Dec 17 2009 9:37 am
From: "Richardson, Patrick"

Well, first of all it doesn't look like you're trying to read an excel file as the file you are trying you specify is "data.txt" not "data.xls".

Try:

library(gdata)
?read.xls

HTH,

Patrick

-----Original Message-----
From: r-help-bounces@r-project.org [mailto:r-help-bounces@r-project.org] On Behalf Of sta_2279@spectrum.net.in
Sent: Thursday, December 17, 2009 12:25 PM
To: r-help@r-project.org
Subject: [R] how to import data from excel to R

Hi,
I am using R and I want to know how data can be transferred from Excel
Spread sheet to R for analyzing. I have done like this

mydata<-read.table("C:\Documents and Settings\admin\Desktop\data.txt");

but its not working how can i do it

regards

Sarath Sankar V
Spectrum Softtech Solution

== 3 of 8 ==
Date: Thurs, Dec 17 2009 9:54 am
From: Joshua Wiley

Hey,

If you want to read it in as a text file, make sure to have Excel save it as
a text file first. For instance, a tab delimited text file. Then you could
read it in with

mydata <- read.table("myfile.txt", sep="\t")

HTH,

Josh

On Thu, Dec 17, 2009 at 9:24 AM, <sta_2279@spectrum.net.in> wrote:

> Hi,
> I am using R and I want to know how data can be transferred from Excel
> Spread sheet to R for analyzing. I have done like this
>
> mydata<-read.table("C:\Documents and Settings\admin\Desktop\data.txt");
>
>
> but its not working how can i do it
>
> regards
>
> Sarath Sankar V
> Spectrum Softtech Solution
>
> ______________________________________________
> R-help@r-project.org mailing list
> https://stat.ethz.ch/mailman/listinfo/r-help
> PLEASE do read the posting guide
> http://www.R-project.org/posting-guide.html<http://www.r-project.org/posting-guide.html>
> and provide commented, minimal, self-contained, reproducible code.
>

--
Joshua Wiley
Senior in Psychology
University of California, Riverside
http://www.joshuawiley.com/

[[alternative HTML version deleted]]

== 4 of 8 ==
Date: Thurs, Dec 17 2009 10:02 am
From: milton ruser

Hi Sarath, try

"C:\\Documents and Settings\\admin\\Desktop\\data.txt"
or
"C:/Documents and Settings/admin/Desktop/data.txt"

or yet

setwd("your path")
mydata<-read.table("data.txt")
cheers

miltinho

On Thu, Dec 17, 2009 at 12:24 PM, <sta_2279@spectrum.net.in> wrote:

[[alternative HTML version deleted]]

== 5 of 8 ==
Date: Thurs, Dec 17 2009 10:35 am
From: David Winsemius

On Dec 17, 2009, at 11:24 AM, sta_2279@spectrum.net.in wrote:

> Hi,
> I am using R and I want to know how data can be transferred from
> Excel
> Spread sheet to R for analyzing. I have done like this
>
> mydata<-read.table("C:\Documents and Settings\admin\Desktop
> \data.txt");
>
>
> but its not working how can i do it

You've gotten three distinct suggestions, all of which are probably
needed and none of which would completely fill in the missing aspects
of your R-knowledge needed for success:

a) reading xls formatted files is possible with quite a few special
functions possibly requiring importing packages.
b) reading exported tab or comma delimited files is possible.
c) your file specification would not have succeeded.
d) you did not report the full error message which suggests the you
should now read the Posting Guide more thoroughly.
e) you should read the documentation in the Import Export Manual.
(This is probably also illustrated in several of the introductory
manuals, and we expect you to read at least one before posting
questions.

Do some homework and then you should have an increase in your store of
knowledge.

--
David.

>
> regards
>
> Sarath Sankar V
> Spectrum Softtech
> Solution
>
> ______________________________________________
> R-help@r-project.org mailing list
> https://stat.ethz.ch/mailman/listinfo/r-help
> PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
> and provide commented, minimal, self-contained, reproducible code.

== 6 of 8 ==
Date: Thurs, Dec 17 2009 11:18 am
From: Gabor Grothendieck

Check out these notes:

http://wiki.r-project.org/rwiki/doku.php?id=tips:data-io:ms_windows

and note that within quotes you must double each backslash so one
writes "c:\\mydirectory\\abc.xls" for example.

On Thu, Dec 17, 2009 at 12:24 PM, <sta_2279@spectrum.net.in> wrote:
> Hi,
> I am using R and I want to know how data can be transferred from Excel
> Spread sheet to R for analyzing. I have done like this
>
> mydata<-read.table("C:\Documents and Settings\admin\Desktop\data.txt");
>
>
> but its not working how can i do it
>
> regards
>
> Sarath Sankar V
> Spectrum Softtech Solution
>
> ______________________________________________
> R-help@r-project.org mailing list
> https://stat.ethz.ch/mailman/listinfo/r-help
> PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
> and provide commented, minimal, self-contained, reproducible code.
>

== 7 of 8 ==
Date: Thurs, Dec 17 2009 12:07 pm
From: Charlie Sharpsteen

On Thu, Dec 17, 2009 at 9:24 AM, <sta_2279@spectrum.net.in> wrote:
>
> Hi,
> I am using R and I want to know how data can be transferred from Excel
> Spread sheet to R for analyzing. I have done like this
>
> mydata<-read.table("C:\Documents and Settings\admin\Desktop\data.txt");
>
>
> but its not working how can i do it

The "R Data Import/Export" manual covers this topic in general:

http://cran.r-project.org/doc/manuals/R-data.html

Chapter 8 concerns getting data into and out of excel. This post on
the Learning R blog also covers many available options for importing
directly from Excel files:

http://learnr.wordpress.com/2009/10/06/export-data-frames-to-multi-worksheet-excel-file/

Hope this helps!

-Charlie

== 8 of 8 ==
Date: Thurs, Dec 17 2009 1:12 pm
From: "Steve Taylor"

Or copy a rectangle of data in Excel, then grab it from the clipboard in R as follows:

my.data <- read.table("clipboard",head=TRUE,sep='\t')
cheers,
Steve

[[alternative HTML version deleted]]

==============================================================================
TOPIC: SPLUS Seqtrial vs. R Packages for sequential clinical trials designs
http://groups.google.com/group/r-help-archive/t/918a8fb554b8161e?hl=en
==============================================================================

== 1 of 2 ==
Date: Thurs, Dec 17 2009 9:34 am
From: David Freedman <3.14david@gmail.com>

You'll probably want to take a look at the CRAN Task View,
http://cran.r-project.org/web/views/ClinicalTrials.html

david freedman

Paul Miller wrote:
>
> Hello Everyone,
> Â
> Iâ€™m a SAS user who has recently become interested in sequential clinical
> trials designs. Iâ€™ve discovered that the SASÂ based approaches for these
> designs are either too costly or areÂ â€œexperimental.â€ So now Iâ€™m
> looking for alternative software. Two programs that seem promising are
> SPLUS Seqtrial and R.
> Â
> I recently obtained a 30 day trial for the SPLUS Seqtrial add-on and have
> worked my way through most of the examples in the manual. Iâ€™ve also
> gotten access to R, installed a package called gsDesign, and worked
> through most of the examples in its documentation.
> Â
> Although I donâ€™t yet have a good understanding of the various approaches
> to sequential clinical trials designs, I thought that the gsDesign package
> seemed very impressive. I also understand that there are several other R
> packages that relate to sequential clinical trials designs, such as
> AGSDest , GroupSeq, ldbounds, MChtest, PwrGSD, and Seqmon. Some of these
> seem fairly comprehensive while others seem to focus on a single approach.
> Â
> My questions center on the adequacy of SPLUS Seqtrial and the R Packages.
> I was wondering if there is anyone out there who would be familiar enough
> with these to comment on their relative merits. Will SPLUS Seqtrial or the
> R packages allow me to do all the designs Iâ€™m ever likely to need? If I
> pay for SPLUS Seqtrial, will I get anything that I canâ€™t get using the
> various R packages? Are any of the R packages comprehensive? Or would it
> at least be possible to cover all the types of designs that are commonly
> used by employing a variety of R packages? What kind of validation work
> generally goes into an R package and how would this likely compare to the
> sort of validation work that has gone into Seqtrial?
> Â
> There may be other questions that I should be asking but havenâ€™t thought
> of.
> Â
> At any rate, if some of you would be willing to share some advice or
> insights Iâ€™d greatly appreciate it.
> Â
> Thanks,
> Â
> Paul
>
>
> __________________________________________________________________
> The new Internet ExplorerÂ® 8 - Faster, safer, easier. Optimized for Y
> er/
> [[alternative HTML version deleted]]
>
>
> ______________________________________________
> R-help@r-project.org mailing list
> https://stat.ethz.ch/mailman/listinfo/r-help
> PLEASE do read the posting guide
> http://www.R-project.org/posting-guide.html
> and provide commented, minimal, self-contained, reproducible code.
>
>

--
View this message in context: http://n4.nabble.com/SPLUS-Seqtrial-vs-R-Packages-for-sequential-clinical-trials-designs-tp966071p966780.html
Sent from the R help mailing list archive at Nabble.com.

== 2 of 2 ==
Date: Thurs, Dec 17 2009 11:13 am
From: Marc Schwartz

On Dec 17, 2009, at 9:58 AM, Paul Miller wrote:

> Hello Everyone,
>
> I'm a SAS user who has recently become interested in sequential
> clinical trials designs. I've discovered that the SAS based
> approaches for these designs are either too costly or are
> "experimental." So now I'm looking for alternative software. Two
> programs that seem promising are SPLUS Seqtrial and R.
>
> I recently obtained a 30 day trial for the SPLUS Seqtrial add-on and
> have worked my way through most of the examples in the manual. I've
> also gotten access to R, installed a package called gsDesign, and
> worked through most of the examples in its documentation.
>
> Although I don't yet have a good understanding of the various
> approaches to sequential clinical trials designs, I thought that the
> gsDesign package seemed very impressive. I also understand that
> there are several other R packages that relate to sequential
> clinical trials designs, such as AGSDest , GroupSeq, ldbounds,
> MChtest, PwrGSD, and Seqmon. Some of these seem fairly comprehensive
> while others seem to focus on a single approach.
>
> My questions center on the adequacy of SPLUS Seqtrial and the R
> Packages. I was wondering if there is anyone out there who would be
> familiar enough with these to comment on their relative merits. Will
> SPLUS Seqtrial or the R packages allow me to do all the designs I'm
> ever likely to need? If I pay for SPLUS Seqtrial, will I get
> anything that I can't get using the various R packages? Are any of
> the R packages comprehensive? Or would it at least be possible to
> cover all the types of designs that are commonly used by employing a
> variety of R packages? What kind of validation work generally goes
> into an R package and how would this likely compare to the sort of
> validation work that has gone into Seqtrial?
>
> There may be other questions that I should be asking but haven't
> thought of.
>
> At any rate, if some of you would be willing to share some advice or
> insights I'd greatly appreciate it.
>
> Thanks,
>
> Paul

In addition to David's pointer to the relevant CRAN Task View, I would
point you to the following document, which is applicable to R ("Base
R" and "Recommended Packages" only):

http://www.r-project.org/doc/R-FDA.pdf

It does not however, apply to user contributed packages (on CRAN or
elsewhere).

When it comes to validation, the majority of the burden is on you and
not on the vendor. The vendor can document their own SDLC and the
document above covers R's, within the regulatory domain of clinical
trials (eg. 21 CFR 11, etc.). However, you cannot validate software in
a vacuum, so you must implement that process as appropriate for your
own shop.

For user contributed packages, you will want to consider communicating
with the package authors/maintainers on their SDLC. Some will no doubt
be better than others and only you can make the risk/benefit
assessment along with your implementing the requisite internal IQ/OQ/
PQ procedures.

From purely a functional comparison perspective, you would need to
list the functionality of the S+ SeqTrial product and then compare it
with the functionality available in R and the user contributed
packages. Only you can make the 'value judgement' as to cost versus
functionality.

It is impossible to project into the future to know what potential
trial designs you may encounter in your work. It is also impossible to
know what new techniques (especially for adaptive designs) may become
available moving forward.

If you search the list archives, using keywords such as SAS or FDA,
you will find a plethora of discussions on the general topics of R
versus SAS, etc. You will also see many discussions on R and clinical
trials, including many of the misconceptions regarding validation and
the FDA (which BTW, uses R internally).

As a SAS user, on a more general level, you may find the following
book to be helpful in moving to R:

R for SAS and SPSS Users
Robert A. Muenchen
http://www.amazon.com/SAS-SPSS-Users-Statistics-Computing/dp/0387094172/

HTH,

Marc Schwartz

==============================================================================
TOPIC: R on amazon's EC2 "cloud"?
http://groups.google.com/group/r-help-archive/t/fd043f2c463ff4ce?hl=en
==============================================================================

== 1 of 3 ==
Date: Thurs, Dec 17 2009 9:55 am
From: Blair Christian

Hi All,

I was wondering if anybody had experience running R on amazon's ec2
"cloud"? More specifically, has anybody created an "amazon machine
image" (AMI) for use with it? I was wondering if there was a quantian
type image available as an amazon machine image, maybe running debian
or ubuntu? It supports open MPI among other libs which is nice.

It would be nice to put together a hardware optimized image that the
community could use (eg lapack with openmp blas support that kind of
thing), maybe a debian with no X installed, but including R + all
packages?

http://aws.amazon.com/ec2/#pricing

I'm at the point where it's almost easier/cost effective to outsource
the hardware...

I bring this up because I saw that they were going to start a spot
market for cycles- useful for things like MCMC chains that need lots
of runs...
http://bit.ly/8A2e1g

== 2 of 3 ==
Date: Thurs, Dec 17 2009 10:03 am
From: Roy Mendelssohn

Go to:

http://biocep-distrib.r-forge.r-project.org/

Choose "documentation", scroll own you will see a section called:

"Biocep-R on Amazon's Cloud"

-roy M.

On Dec 17, 2009, at 9:55 AM, Blair Christian wrote:

> Hi All,
>
> I was wondering if anybody had experience running R on amazon's ec2
> "cloud"? More specifically, has anybody created an "amazon machine
> image" (AMI) for use with it? I was wondering if there was a quantian
> type image available as an amazon machine image, maybe running debian
> or ubuntu? It supports open MPI among other libs which is nice.
>
> It would be nice to put together a hardware optimized image that the
> community could use (eg lapack with openmp blas support that kind of
> thing), maybe a debian with no X installed, but including R + all
> packages?
>
> http://aws.amazon.com/ec2/#pricing
>
> I'm at the point where it's almost easier/cost effective to outsource
> the hardware...
>
> I bring this up because I saw that they were going to start a spot
> market for cycles- useful for things like MCMC chains that need lots
> of runs...
> http://bit.ly/8A2e1g
>
> ______________________________________________
> R-help@r-project.org mailing list
> https://stat.ethz.ch/mailman/listinfo/r-help
> PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
> and provide commented, minimal, self-contained, reproducible code.

**********************
"The contents of this message do not reflect any position of the U.S. Government or NOAA."
**********************
Roy Mendelssohn
Supervisory Operations Research Analyst
NOAA/NMFS
Environmental Research Division
Southwest Fisheries Science Center
1352 Lighthouse Avenue
Pacific Grove, CA 93950-2097

e-mail: Roy.Mendelssohn@noaa.gov (Note new e-mail address)
voice: (831)-648-9029
fax: (831)-648-8440
www: http://www.pfeg.noaa.gov/

"Old age and treachery will overcome youth and skill."
"From those who have been given much, much will be expected"

== 3 of 3 ==
Date: Thurs, Dec 17 2009 10:15 am
From: Dirk Eddelbuettel

On 17 December 2009 at 12:55, Blair Christian wrote:
| Hi All,
|
| I was wondering if anybody had experience running R on amazon's ec2
| "cloud"? More specifically, has anybody created an "amazon machine
| image" (AMI) for use with it? I was wondering if there was a quantian
| type image available as an amazon machine image, maybe running debian
| or ubuntu? It supports open MPI among other libs which is nice.
|
| It would be nice to put together a hardware optimized image that the
| community could use (eg lapack with openmp blas support that kind of
| thing), maybe a debian with no X installed, but including R + all
| packages?

Ubuntu has been supporting AMIs for a while, one decent link may be

https://help.ubuntu.com/community/EC2StartersGuide

The most recent Ubuntu release contains the REvolution R + Canonical
cooperation comprising the Intel MKLs which should be faster than Atlas.

Dirk

--
Three out of two people have difficulties with fractions.

==============================================================================
TOPIC: user-written splits in rpart
http://groups.google.com/group/r-help-archive/t/d30b1412af8a4672?hl=en
==============================================================================

== 1 of 1 ==
Date: Thurs, Dec 17 2009 11:15 am
From: "Verspagen B (ALGEC)"

--- begin original message ---
I am trying to write my own split function for rpart. The aim is to do,
instead of anova, a linear regression to determine the split (minimize
some criterion like sum of rss left and right of the split). The
regression (lm) should simply use the dependent and independent
variables passed to rpart.

I am aware of the example provided in the rpart source code, but
stumbled on similar problems that I saw reported on this list (no final
solution posted, as far as I could see). The problem is, broadly
speaking, that I do not see a way to access the full set of x and y
variables in the user-written split-function.
---- end original message -----------

---- begin reply -----------
The rpart routine provides the x variables to a user-written split
function one at a time. Since the entire structure of rpart --
printing, plotting, tree representation, etc --- is based on the premise
of a single variable driving each split, what you are asking for would
require an entirely different program.
---- end reply -----------

I have been reflecting a little bit on this. Perhaps I have been unclear.
It is fine with me that a single variable "drives" the split, i.e., that we partition the data based on the ranking of a single variable. I am not looking to change that.
What I would like to change is the evaluation of that split. In the anova implementation of rpart, this evaluation is now done by taking the average of the dependent variable (left and right of the split) as a predictor. Instead of that, I would like to run a linear regression (left and right of the split) involving all x-variabels that were supplied in the call to rpart, and take the predictive power of those regressions as the evaluation of the split.
I can't see how that would change the structure of the entire module.

Best regards,

Bart Verspagen

==============================================================================
TOPIC: Remove duplicates from a data frame but with some special requirements
http://groups.google.com/group/r-help-archive/t/8560b71c8fd887b8?hl=en
==============================================================================

== 1 of 1 ==
Date: Thurs, Dec 17 2009 11:31 am
From: gcam

Thanks Gray,

This helps, I'd completely forgotten about the subset command. However, it
doesn't quite get me where I need. Perhaps an example will help. I will
simplify my dataframe to the three important variables:

ESR_ref ESR_ref_edit Loaded
1.1 1.1 Y
1.1.1 1.1 NC
1.1.2 1.1 Y
2.1 2.1 N
2.1.1 2.1 Y
2.1.2 2.1 PU
2.1.3 2.1 Y
3.1 3.1 Y
4.1 4.1 N
4.1.1 4.1 PU

So I've created the "edit" variable so I can test for duplicates (i.e.
samples with more than one sub-sample) because this is not of interest at
this point. I just want one subsample per sample. However, if we consider
2.1 - this would result in a subset (if duplicates were removed) with the
first line which has an "N". But it is of interest to me the if at least
one of the subsamples has a "Y" then I want that line rather than a
subsample with another code. 1.1 in this example works by default because
the first subsample is a "Y" so it will retain that information.

Thanks

Gareth

Gray Calhoun-2 wrote:
>
> Hi,
> Try:
>
> subset(Samps, !duplicated(Samps$ESR_ref_edit) | Samps$Loaded == "Y")
>
> I'd need specific code to be sure that this is exactly what you want
> (ie you specify input and desired output), but indexing with a logical
> vector is probably going to be the solution.
>
> Best,
> Gray
>
> On Wed, Dec 16, 2009 at 7:55 PM, gcam <gcam032@gmail.com> wrote:
>>
>> Hi all.
>>
>> So I have a data frame with multiple columns/variables. The first
>> variable
>> is a major sample name for which there are some sub-samples. Currently I
>> have used the following command to remove the duplicates:
>>
>> Samps_working<-Samps[-c(which(duplicated(Samps$ESR_Ref_edit))),]
>>
>> This removes all of the duplicated sample rows.
>>
>> However, I just realised that, of course, this removes the first
>> observation
>> of each duplicated set. However, I wish to retain any that have the code
>> "Y" in another variable Samps$Loaded. I'm at a bit of a loss as to how
>> best
>> to approach this problem.
>>
>> Just to reiterate. I want to remove all duplicate lines based on sample
>> name, but, I want the lines to be removed with a preference given to
>> those
>> that do not include a "Y" in the Loaded variable column.
>> --
>> View this message in context:
>> http://n4.nabble.com/Remove-duplicates-from-a-data-frame-but-with-some-special-requirements-tp965745p965745.html
>> Sent from the R help mailing list archive at Nabble.com.
>>
>> ______________________________________________
>> R-help@r-project.org mailing list
>> https://stat.ethz.ch/mailman/listinfo/r-help
>> PLEASE do read the posting guide
>> http://www.R-project.org/posting-guide.html
>> and provide commented, minimal, self-contained, reproducible code.
>>
>
>
>
> --
> Gray Calhoun
>
> Assistant Professor of Economics
> Iowa State University
>
> ______________________________________________
> R-help@r-project.org mailing list
> https://stat.ethz.ch/mailman/listinfo/r-help
> PLEASE do read the posting guide
> http://www.R-project.org/posting-guide.html
> and provide commented, minimal, self-contained, reproducible code.
>
>

--
View this message in context: http://n4.nabble.com/Remove-duplicates-from-a-data-frame-but-with-some-special-requirements-tp965745p974312.html
Sent from the R help mailing list archive at Nabble.com.

==============================================================================
TOPIC: forecasting
http://groups.google.com/group/r-help-archive/t/c3d48bf5e28f6075?hl=en
==============================================================================

== 1 of 2 ==
Date: Thurs, Dec 17 2009 11:50 am
From: DispersionMap

I have some data that i ran a regression on and got the usual r output, with
the a intercept and b coefficient for my independent variable.

i want to forecast the number of future events using these parameters.

What function / packages in R would you recommend to be used in good effect
for these purposes??
--
View this message in context: http://n4.nabble.com/forecasting-tp974326p974326.html
Sent from the R help mailing list archive at Nabble.com.

== 2 of 2 ==
Date: Thurs, Dec 17 2009 11:56 am
From: Stephan Kolassa

?predict

HTH
Stephan

DispersionMap schrieb:
>
> I have some data that i ran a regression on and got the usual r output, with
> the a intercept and b coefficient for my independent variable.
>
> i want to forecast the number of future events using these parameters.
>
>
> What function / packages in R would you recommend to be used in good effect
> for these purposes??

==============================================================================
TOPIC: Sweave Makefile issue
http://groups.google.com/group/r-help-archive/t/e748dd73b2868f5a?hl=en
==============================================================================

== 1 of 2 ==
Date: Thurs, Dec 17 2009 12:19 pm
From: Luc Villandré

Dear R-specialists,

I am trying to create a Makefile that will first convert all my .Rnw
files into .tex files and then, that will run the LaTeX compiler to
produce a pdf document. This issue has been discussed before. Hence,
I've basically adapted a Makefile I found at
http://n4.nabble.com/R-Sweave-R-and-complex-latex-projects-td810020.html#a810023
to make it compatible with a Windows XP environment. You will find my
version of it at the end of this message.

However, it's not doing what I expect it to do. Instead of only using
pdfTeX on "total_article.tex" (which is the file with the necessary
headers and \include's), it tries to convert all my tex files into pdf,
which of course cannot be done since only total_article.tex has the
necessary structure to be understood by the compiler.

What's wrong then with this Makefile (I am using GNU Make 3.81)?

I will be grateful for any help you can provide.
_________________________________________________

MASTER = total_article.pdf

# the master document depends on all of the tex files
Rfile = prepareScript.R
RNWFILES = $(wildcard *.Rnw)
TEXFILES = $(wildcard *.tex)
DEPENDS = $(patsubst %.Rnw,%.tex,$(RNWFILES)) $(TEXFILES)

all : $(MASTER)

$(MASTER) : $(DEPENDS)

%.tex : %.Rnw

SWEAVE '$<'

%.pdf : %.tex

@pdflatex $<
@egrep -c $(RERUNBIB) $*.log && (bibtex $*;pdflatex $<); true
@egrep $(RERUN) $*.log && (pdflatex $<) ; true
@egrep $(RERUN) $*.log && (pdflatex $<) ; true

clean:
@del *.aux *.log *.bbl *.blg *.brf *.cb *.ind *.idx *.ilg \
*.inx *.ps *.dvi *.toc *.out *.lot *~ *.lof *.ttt *.fff \
*.eps *.pdf
@del -f $(patsubst %.Rnw,%.tex,$(RNWFILES))

script :
Rcmd BATCH $(Rfile)

*Luc Villandré*
/Biostatistician
MUHC-MCH Research Institute/

== 2 of 2 ==
Date: Thurs, Dec 17 2009 1:13 pm
From: Gray Calhoun

Hi Luc,

I haven't been actively participating in this list for that long, so
I may be wrong here... but I suspect that debugging makefiles is a bit
off topic. Two quick points that might help: You probably want to use
"R CMD texi2dvi" on your tex files instead of pdflatex; and you'll
probably have better luck finding the dependency error in your
Makefile if you remove all of the variables (especially with
$(wildcard ...)) and create a simplistic directory structure. Once
you get that working, you can build back up. Using Sweave generated
LaTeX files as dependencies for other LaTeX files may be problematic
even when you get this issue sorted out (I'm not familiar enough with
Sweave to know if you can strip out the preamble, etc.)

Best,
Gray

On Thu, Dec 17, 2009 at 2:19 PM, Luc Villandré
<villandl@dms.umontreal.ca> wrote:
> Dear R-specialists,
>
> I am trying to create a Makefile that will first convert all my .Rnw files
> into .tex files and then, that will run the LaTeX compiler to produce a pdf
> document. This issue has been discussed before. Hence, I've basically
> adapted a Makefile I found at
> http://n4.nabble.com/R-Sweave-R-and-complex-latex-projects-td810020.html#a810023
> to make it compatible with a Windows XP environment. You will find my
> version of it at the end of this message.
>
> However, it's not doing what I expect it to do. Instead of only using pdfTeX
> on "total_article.tex" (which is the file with the necessary headers and
> \include's), it tries to convert all my tex files into pdf, which of course
> cannot be done since only total_article.tex has the necessary structure to
> be understood by the compiler.
>
> What's wrong then with this Makefile (I am using GNU Make 3.81)?
>
> I will be grateful for any help you can provide.
> _________________________________________________
>
> MASTER = total_article.pdf
>
> # the master document depends on all of the tex files
> Rfile = prepareScript.R
> RNWFILES = $(wildcard *.Rnw)
> TEXFILES = $(wildcard *.tex)
> DEPENDS = $(patsubst %.Rnw,%.tex,$(RNWFILES)) $(TEXFILES)
>
> RERUN = "(There were undefined references|Rerun to get
> (citations|cross-references|the bars) (correct|right)|Table widths have
> changed. Rerun LaTeX.|Linenumber reference failed)"
> RERUNBIB = "No file.*\.bbl|Citation.*undefined"
>
> all : $(MASTER)
>
> $(MASTER) : $(DEPENDS)
>
> %.tex : %.Rnw
>
> SWEAVE '$<'
>
> %.pdf : %.tex
>
> @pdflatex $<
> @egrep -c $(RERUNBIB) $*.log && (bibtex $*;pdflatex $<); true
> @egrep $(RERUN) $*.log && (pdflatex $<) ; true
> @egrep $(RERUN) $*.log && (pdflatex $<) ; true
>
> clean:
> @del *.aux *.log *.bbl *.blg *.brf *.cb *.ind *.idx *.ilg \
> *.inx *.ps *.dvi *.toc *.out *.lot *~ *.lof *.ttt *.fff \
> *.eps *.pdf
> @del -f $(patsubst %.Rnw,%.tex,$(RNWFILES))
> script :
> Rcmd BATCH $(Rfile)
>
> --
>
> *Luc Villandré*
> /Biostatistician
> MUHC-MCH Research Institute/
>
> ______________________________________________
> R-help@r-project.org mailing list
> https://stat.ethz.ch/mailman/listinfo/r-help
> PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
> and provide commented, minimal, self-contained, reproducible code.
>

--
Gray Calhoun

Assistant Professor of Economics
Iowa State University

==============================================================================
TOPIC: Problem reading binaries created with fortran
http://groups.google.com/group/r-help-archive/t/25da4b41f780a4e8?hl=en
==============================================================================

== 1 of 3 ==
Date: Thurs, Dec 17 2009 12:48 pm
From: kapo coulibaly

Is it possible to read fortran binaries with R? I tried unsucessfully and my
understanding is that fortran write binaries with leading and trailing
bytes. I get numbers but not the right ones.
Thanks

ps: the binary I'm interested in reading is a MODFLOW output with a mix of
character, double and integers.

[[alternative HTML version deleted]]

== 2 of 3 ==
Date: Thurs, Dec 17 2009 1:00 pm
From: Duncan Murdoch

On 17/12/2009 3:48 PM, kapo coulibaly wrote:
> Is it possible to read fortran binaries with R? I tried unsucessfully and my
> understanding is that fortran write binaries with leading and trailing
> bytes. I get numbers but not the right ones.
> Thanks
>
> ps: the binary I'm interested in reading is a MODFLOW output with a mix of
> character, double and integers.

R can read most binary files, as long as you tell it the right format to
read. You need to consult the documentation of the program that
produced the file to find out the format. If it adds leading or
trailing bytes, just tell R to skip over those --- but you need to know
how much to skip.

Duncan Murdoch

== 3 of 3 ==
Date: Thurs, Dec 17 2009 1:02 pm
From: Duncan Murdoch

One other thing: the hexView package does a very nice job of displaying
the file, so you can work out what the structure is if the documentation
is unclear (or nonexistent).

Duncan Murdoch

==============================================================================
TOPIC: ROracle on Windows Vista?
http://groups.google.com/group/r-help-archive/t/2b6847ef791578a4?hl=en
==============================================================================

== 1 of 1 ==
Date: Thurs, Dec 17 2009 1:09 pm
From: cstrato

Dear All,

Does anybody know whether the package ROracle_0.5-9 does compile and run
on Windows Vista?
(Since I do not have Oracle I am not able to test this myself.)

I have tried to contact the maintainer of ROracle but it seems that the
email address is no longer valid.

Best regards
Christian
_._._._._._._._._._._._._._._._._._
C.h.r.i.s.t.i.a.n S.t.r.a.t.o.w.a
V.i.e.n.n.a A.u.s.t.r.i.a
e.m.a.i.l: cstrato at aon.at
_._._._._._._._._._._._._._._._._._

==============================================================================

You received this message because you are subscribed to the Google Groups "R-help-archive"
group.

To post to this group, send email to r-help@r-project.org or visit http://groups.google.com/group/r-help-archive?hl=en

To unsubscribe from this group, send email to r-help-archive+unsubscribe@googlegroups.com

To change the way you get mail from this group, visit:
http://groups.google.com/group/r-help-archive/subscribe?hl=en

To report abuse, send email explaining the problem to abuse@googlegroups.com

==============================================================================
Google Groups: http://groups.google.com/?hl=en

r-help-archive - 25 new messages in 13 topics - digest

2009-12-17T08:01:00.000-08:00

R-help-archive
http://groups.google.com/group/r-help-archive?hl=en

r-help-archive@googlegroups.com

Today's topics:

* Random Number Generation in a Loop - 1 messages, 1 author
http://groups.google.com/group/r-help-archive/t/79fc3ec13999b575?hl=en
* Extract variance from dist(cluster.data) - 1 messages, 1 author
http://groups.google.com/group/r-help-archive/t/e6b3b8b805e8bbcd?hl=en
* Error for making packages under windows XP-Error in library.dynam(lib,
package, package.lib) - 1 messages, 1 author
http://groups.google.com/group/r-help-archive/t/44f9d82095e23cf3?hl=en
* CORRECTION - Generation of Random numbers in a loop - 1 messages, 1 author
http://groups.google.com/group/r-help-archive/t/d548cf46d887fbeb?hl=en
* What is the fastest way to see what are in an RData file? - 2 messages, 2
authors
http://groups.google.com/group/r-help-archive/t/f70fcf5f3d959ef0?hl=en
* issue with using rm: cannot generate on-the-fly list - 3 messages, 3 authors
http://groups.google.com/group/r-help-archive/t/c9724d17a9c0487e?hl=en
* Fishers exact test at < 2.2e-16 - 5 messages, 5 authors
http://groups.google.com/group/r-help-archive/t/f57ef044081517d8?hl=en
* Problem with spliting a dataframe values - 2 messages, 2 authors
http://groups.google.com/group/r-help-archive/t/a8145570b96c7f67?hl=en
* nonlinear (especially logistic) regression accounting for spatially
correlated errors - 1 messages, 1 author
http://groups.google.com/group/r-help-archive/t/31a97437c19744d4?hl=en
* Exchange NAs for mean - 4 messages, 4 authors
http://groups.google.com/group/r-help-archive/t/c61616a48d5610a0?hl=en
* unable to run JGR in windows 7 - 1 messages, 1 author
http://groups.google.com/group/r-help-archive/t/d802a6dcbd5be27d?hl=en
* cross-product - 2 messages, 2 authors
http://groups.google.com/group/r-help-archive/t/853af62ce8a99bf4?hl=en
* SPLUS Seqtrial vs. R Packages for sequential clinical trials designs - 1
messages, 1 author
http://groups.google.com/group/r-help-archive/t/918a8fb554b8161e?hl=en

==============================================================================
TOPIC: Random Number Generation in a Loop
http://groups.google.com/group/r-help-archive/t/79fc3ec13999b575?hl=en
==============================================================================

== 1 of 1 ==
Date: Wed, Dec 16 2009 11:23 pm
From: "Daniel Malter"

Hi, does this do what you want to do? I take the first six lines of the data
you provided. Note that "letters" corresponds to your "Name," and "number"
corresponds to your "Numbers" variable.

letters=rep(c("A","B","C"),2)
letters

number=c(25,3,13,5,7,0)
number

letters2=rep(letters,number)
letters2

interval.min=ifelse(letters2=="A",1.05,ifelse(letters2=="B",1.3,1.6))
interval.max=ifelse(letters2=="A",1.3,ifelse(letters2=="B",1.6,1.99))

rand.nums=runif(letters2,min=interval.min,max=interval.max)

data.frame(letters2,rand.nums)

HTH,
Daniel

-------------------------
cuncta stricte discussurus
-------------------------
-----Original Message-----
From: r-help-bounces@r-project.org [mailto:r-help-bounces@r-project.org] On
Behalf Of Maithili Shiva
Sent: Thursday, December 17, 2009 1:25 AM
To: r-help@r-project.org
Subject: [R] Random Number Generation in a Loop

Dear R helpers

I am having following data

Name Numbers
A 25
B 3
C 13
A 5
B 7
C 0
A 2
B 10
C 9

CONDITIONS

If Name is A, min_val = 1.05, max_val = 1.30 If Name is B, min_val = 1.30,
max_val = 1.60 If Name is C, min_val = 1.60, max_val = 1.99

TASK

To generate the Uniform random nos for each of these Names (Equal to the
corresponding no. e.g. the 5th Name is B, so I need to generate 7 uniform
random numbers in the range (1.30 - 1.99). Also I need to arrange these
random numbers one by one in a single csv file i.e. say 25 random numbers in
teh range (1.05-1.30) followed by 3 random numbers in the range (1.30-1.60)
and so on.

# ___________________________________________________

Here is the R code I have tried

ONS <- read.table(textConnection("name number
A11 12
A12 17
A13 0
A11 11
A12 6
A13 0
A11 8
A12 4
A13 3"), header = TRUE)

X = as.character(ONS$name)
Y = ONS$number

Z = NULL

for (i in 1:length(X))
{
if(X[i] == 'A11')
{
min_val = 1.05
max_val = 1.30
Z = runif(Y[i], min_val, max_val)
}
else
{
if(X[i] == 'A12')
{
min_val = 1.30
max_val = 1.60
Z = runif(Y[i], min_val, max_val)
}
else
{
if(X[i] == 'A13')
{
min_val = 1.60
max_val = 1.99
Z = runif(Y[i], min_val, max_val)
}
}
}
}

# End of Code

## _____________________________________________________________________

PROBLEM

I need to get 61 random numbers which is total of all the numbers (1st 12 fo
A, 3 random numbers for B, 13 for C, 5 again fo A and so on). The result
whcih I got is

> Z
[1] 1.740443 1.761758 1.797222

which is pertaining to the last name C where 3 random numbers are generated
i.e. Z instaed of getting added, is overwritten.

Please help me to rectify my code so that in the end I will get 61 random
numbers as desired i.e. 12 for A in the range (1.05 - 1.30), 3 for B in the
range (1.30 - 1.60), 13 for C in the range (1.60-1.99), again 5 for A in the
range (1.05 - 1.30).

Thanking in advance. I also sincerely apologize for writing such a long
mail, as I wanted to be clear as possible in my communication.

Regards

Maithili

The INTERNET now has a personality. YOURS! See your Yahoo! Homepage.
[[alternative HTML version deleted]]

==============================================================================
TOPIC: Extract variance from dist(cluster.data)
http://groups.google.com/group/r-help-archive/t/e6b3b8b805e8bbcd?hl=en
==============================================================================

== 1 of 1 ==
Date: Thurs, Dec 17 2009 1:00 am
From: "Philip Silva"

Hi!

I am currently clustering data that is read out of a csv file. The code is:

file <- read.table("/tmp/cluster_input.tmp")
cluster.data <- as.matrix(file[,2:39040])
cluster.dist <- dist(cluster.data)
cluster.hclust <- hclust(cluster.dist)
cluster.hcut <- cutree(cluster.hclust,10) # 10 == number of clusters!
write.table(cluster.hcut, "/tmp/cluster_result.csv")

Now I would like to know the Variance of each cluster in hcut. Is it correct that the variances are already encoded in dist(cluster.data)? How can I get it?

Thanks a lot in advance.

Best regards
Philip Silva
--
Preisknaller: GMX DSL Flatrate für nur 16,99 Euro/mtl.!
http://portal.gmx.net/de/go/dsl02

== 1 of 1 ==
Date: Thurs, Dec 17 2009 3:14 am
From: Duncan Murdoch

rusers.sh wrote:
> Hi,
> I have installed the necessary tools for making a R package under windows
> and am sure these tool have been correctly configured. I am very new to try
> writing a package, so the error may be very obvious for you. Forgive me if
> it is too easy.
> My package name is *stam*, and i put it under the folder
> "D:/StatSoft/R/MyPackage/". I have two folders,*R* and *man*. Besides, i
> also have *description* and *namespace** *files. It is a very simple beta
> package.
>

The DESCRIPTION and NAMESPACE files should be specified in upper case,
but that's probably not your problem. I would guess you have a command
that says to load a shared library, e.g. a useDynLib statement in your
NAMESPACE, or a library.dynam in a .First() function. Just delete it,
without compiled code (which would be kept in the src directory) you
don't need it.

Duncan Murdoch
> Following is the error after using "R CMD check mypackage". Thanks very
> much.
> #########
> * using log directory 'D:/StatSoft/R/MyPackage/stam.Rcheck'
> * using R version 2.10.0 (2009-10-26)
> * using session charset: ISO8859-1
> * checking for file 'stam/DESCRIPTION' ... OK
> * this is package 'stam' version '1.0-0'
> * checking package name space information ... OK
> * checking package dependencies ... OK
> * checking if this is a source package ... OK
> * checking for executable files ... OK
> * checking whether package 'stam' can be installed ... OK
> * checking package directory ... OK
> * checking for portable file names ... OK
> * checking DESCRIPTION meta-information ... OK
> * checking top-level files ... OK
> * checking index information ... OK
> * checking package subdirectories ... OK
> * checking R files for non-ASCII characters ... OK
> * checking R files for syntax errors ... OK
> * checking whether the package can be loaded ... ERROR
> Error in library.dynam(lib, package, package.lib) :
> shared library 'stam' not found
> Error: package/namespace load failed for 'stam'
> Execution halted
>
> It looks like this package has a loading problem: see the messages for
> details.
>
> #########
>
>

==============================================================================
TOPIC: CORRECTION - Generation of Random numbers in a loop
http://groups.google.com/group/r-help-archive/t/d548cf46d887fbeb?hl=en
==============================================================================

== 1 of 1 ==
Date: Thurs, Dec 17 2009 3:27 am
From: Gray Calhoun

Hi Maithili,
This is probably easiest if you think in terms of functions. Just
apply 'runif' along the right vectors

## gives you one data frame with the min and max
limits <- read.table(textConnection("name min max
A11 1.05 1.30
A12 1.30 1.60
A13 1.60 1.99"), header = TRUE)
d <- merge(ONS, limits)

cbind(randomdraw = unlist(mapply(runif, d$number, d$min, d$max)), name
= rep(d$name, d$number))

## now save it to a csv file, etc.

On Wed, Dec 16, 2009 at 10:35 PM, Maithili Shiva
<maithili_shiva@yahoo.com> wrote:
> Dear R helpers, please ignore my earlier mail. Here is the corrected mail. Please forgive me for the lapses on my part. Extremely sorry.
>
> Here is the corrected mail.
>
>
> Dear R helpers
>
> I am having following data
>
> Name           Numbers
> A11                  12
> A12                  17
> A13                   0
> A11                  11
> A12                   6
> A13                   0
> A11                   8
> A12                   4
> A13                   3
>
> CONDITIONS
>
> If Name is A11, min_val = 1.05, max_val = 1.30
> If Name is A12, min_val = 1.30, max_val = 1.60
> If Name is A13, min_val = 1.60, max_val = 1.99
>
> TASK
>
> To generate the Uniform random nos for each of these Names (Equal to the corresponding no. e.g. the 5th Name is A12, so I need to generate 6 uniform random numbers in the range (1.30 - 1.99). Also I need to arrange these random numbers one by one in a single csv file i.e. say 12 random numbers in teh range (1.05-1.30) followed by 17 random numbers in the range (1.30-1.60) and so on.
>
> # ___________________________________________________
>
> Here is the R code I have tried
>
> ONS <- read.table(textConnection("name number
> A11    12
> A12    17
> A13     0
> A11    11
> A12      6
> A13      0
> A11      8
> A12      4
> A13      3"), header = TRUE)
>
> X = as.character(ONS$name)
> Y = ONS$number
>
> Z = NULL
>
> for (i in 1:length(X))
>    {
>    if(X[i] == 'A11')
>    {
>    min_val = 1.05
>    max_val = 1.30
>    Z = runif(Y[i], min_val, max_val)
>    }
>    else
>    {
>    if(X[i] == 'A12')
>    {
>    min_val = 1.30
>    max_val = 1.60
>    Z = runif(Y[i], min_val, max_val)
>    }
>    else
>    {
>    if(X[i] == 'A13')
>    {
>    min_val = 1.60
>    max_val = 1.99
>    Z = runif(Y[i], min_val, max_val)
>    }
>    }
>    }
>    }
>
> # End of Code
>
> ## _____________________________________________________________________
>
> PROBLEM
>
> I need to get 61 random numbers which is total of all the numbers (1st 12 fo A, 3 random numbers for B, 13 for C, 5 again fo A and so on). The result whcih I got is
>
>> Z
> [1] 1.740443 1.761758 1.797222
>
> which is pertaining to the last name C where 3 random numbers are generated i.e. Z instaed of getting added, is overwritten.
>
> Please help me to rectify my code so that in the end I will get 61 random numbers as desired i.e. 12 for A in the range (1.05 - 1.30), 3 for B in the range (1.30 - 1.60), 13 for C in the range (1.60-1.99), again 5 for A in the range (1.05 - 1.30).
>
> Thanking in advance. I also sincerely apologize for writing such a long mail, as I wanted to be clear as possible in my communication.
>
> Regards
>
> Maithili
>
>
> The INTERNET now has a personality. YOURS! See your Yahoo! Homepage.
> [[alternative HTML version deleted]]
>
>
> ______________________________________________
> R-help@r-project.org mailing list
> https://stat.ethz.ch/mailman/listinfo/r-help
> PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
> and provide commented, minimal, self-contained, reproducible code.
>
>

--
Gray Calhoun

Assistant Professor of Economics
Iowa State University

==============================================================================
TOPIC: What is the fastest way to see what are in an RData file?
http://groups.google.com/group/r-help-archive/t/f70fcf5f3d959ef0?hl=en
==============================================================================

== 1 of 2 ==
Date: Thurs, Dec 17 2009 3:33 am
From: Gustaf Rydevik

On Wed, Dec 16, 2009 at 10:13 PM, Peng Yu <pengyu.ut@gmail.com> wrote:
>
> Currently, I load the RData file then ls() and str(). But loading the file
> takes too long if the file is big. Most of the time, I only interested what
> the variables are in the the file and the attributes of the variables (like
> if it is a data.frame, matrix, what are the colnames/rownames, etc.)
>
> I'm wondering if there is any facility in R to help me avoid loading the
> whole file.

I thought this was interesting as well, so i did a bit of searching
through the R-help list archives and found this answer by Simon
Urbanek:
https://stat.ethz.ch/pipermail/r-devel/2007-August/046724.html
The link to a c-routine that does what you want still works, but for
future reference I'm pasting the code below.

Regards,
Gustaf

----------------------------
/* rdcopy v0.1-0 - extract objects or display contents of RData RDX2 files
*
* Copyright (C) 2007 Simon Urbanek
* based in part on src/main/serialize.c and src/main/saveload.c from R:
* Copyright (C) 1995, 1996 Robert Gentleman and Ross Ihaka
* Copyright (C) 1997--2007 Robert Gentleman, Ross Ihaka and the
* R Development Core Team
* License: GPL v2
*
* Although R includes are needed to compile this (for constants),
* libR does NOT have to be linked.
*/

#include <stdio.h>
#include <rpc/types.h>
#include <rpc/xdr.h>
#include <R.h>
#include <Rinternals.h>

#ifndef _
#define _(X) X

r-help-archive - 7 new messages in 5 topics - digest

2009-12-17T00:48:00.001-08:00

R-help-archive
http://groups.google.com/group/r-help-archive?hl=en

r-help-archive@googlegroups.com

Today's topics:

* Remove duplicates from a data frame but with some special requirements - 1
messages, 1 author
http://groups.google.com/group/r-help-archive/t/8560b71c8fd887b8?hl=en
* Help with Merge - unexpected loss of factor level - 2 messages, 2 authors
http://groups.google.com/group/r-help-archive/t/c542f135db6b5bd6?hl=en
* change the scale of grid lines in a plot - 1 messages, 1 author
http://groups.google.com/group/r-help-archive/t/ed2627a0ec1bf29c?hl=en
* Random Number Generation in a Loop - 2 messages, 2 authors
http://groups.google.com/group/r-help-archive/t/79fc3ec13999b575?hl=en
* CORRECTION - Generation of Random numbers in a loop - 1 messages, 1 author
http://groups.google.com/group/r-help-archive/t/d548cf46d887fbeb?hl=en

== 1 of 1 ==
Date: Wed, Dec 16 2009 9:08 pm
From: Gray Calhoun

Hi,
Try:

subset(Samps, !duplicated(Samps$ESR_ref_edit) | Samps$Loaded == "Y")

I'd need specific code to be sure that this is exactly what you want
(ie you specify input and desired output), but indexing with a logical
vector is probably going to be the solution.

Best,
Gray

On Wed, Dec 16, 2009 at 7:55 PM, gcam <gcam032@gmail.com> wrote:
>
> Hi all.
>
> So I have a data frame with multiple columns/variables. The first variable
> is a major sample name for which there are some sub-samples. Currently I
> have used the following command to remove the duplicates:
>
> Samps_working<-Samps[-c(which(duplicated(Samps$ESR_Ref_edit))),]
>
> This removes all of the duplicated sample rows.
>
> However, I just realised that, of course, this removes the first observation
> of each duplicated set. However, I wish to retain any that have the code
> "Y" in another variable Samps$Loaded. I'm at a bit of a loss as to how best
> to approach this problem.
>
> Just to reiterate. I want to remove all duplicate lines based on sample
> name, but, I want the lines to be removed with a preference given to those
> that do not include a "Y" in the Loaded variable column.
> --
> View this message in context: http://n4.nabble.com/Remove-duplicates-from-a-data-frame-but-with-some-special-requirements-tp965745p965745.html
> Sent from the R help mailing list archive at Nabble.com.
>
> ______________________________________________
> R-help@r-project.org mailing list
> https://stat.ethz.ch/mailman/listinfo/r-help
> PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
> and provide commented, minimal, self-contained, reproducible code.
>

--
Gray Calhoun

Assistant Professor of Economics
Iowa State University

==============================================================================
TOPIC: Help with Merge - unexpected loss of factor level
http://groups.google.com/group/r-help-archive/t/c542f135db6b5bd6?hl=en
==============================================================================

== 1 of 2 ==
Date: Wed, Dec 16 2009 9:17 pm
From: Zoe van Havre

Hi, Thanks in advance for any advice you can give me, I am very stumped on this problem...

I use R every day and consider myself a confident user, but this seems to be an elementary problem..

Outline of problem: I am analysing the results of a study on protein expression in cancer tissues. I have raw intensities from 2 different types of cancer and normal tissue, which can be taken from several different parts of the cell, as well as patient information. Part of the analysis calls for a fold-change calculation. In order to do this I am sub-setting the dataset by cancer type, merging each cancer dataset with the data from the Normal tissue, then calculating fold change for matching individuals and cell section.

The problem is that I have been tracking one factor in particular ('branch', values 2 or 3) and once the final merge occurs, the second level of this factor seems to disappear in the last dataset, even though it was present before.
See code & output below:

> dim(tma)
> names(tma)
[1] "Code" "marker" "cell" "tumourA" "tumourEXP" "int" "stain" "tumourPERC" "branch"
> levels(tma$tumourA)
[1] "DCIS" "LN Metastasis" "Normal" "Primary Invasive Carcinoma"
#split into cancer and normal tissue
> tma1<-subset(tma, tumourA=="Primary Invasive Carcinoma")
> tma2<-subset(tma, tumourA=="LN Metastasis")
> tmaN<-subset(tma, tumourA=="Normal")

#size of datasets
> dim(tma1)
[1] 587 9
> dim(tma2)
[1] 323 9
> dim(tmaN)
[1] 142 9

#merge back with normal type
> tma1.1<-merge(tmaN, tma1, by="Code")
> tma2.1<-merge(tmaN, tma2, by="Code")

#new dimensions (seem excessively large)
> dim(tma1.1)
[1] 2439 17
> dim(tma2.1)
[1] 625 17

#progression of "branch: factor in datasets. Note last one where it disappears...
> table(tma$branch)

2 3
450 613
> table(tma1$branch)

2 3
314 273
> table(tma2$branch)

2 3
39 284
> table(tmaN$branch)

2 3
91 51
> table(tma1.1$branch.x)

2 3
1806 633
> table(tma2.1$branch.x)

3
625

Please, can someone tell me what's going on?

Thanks you very much,
Zoe van Havre

[[alternative HTML version deleted]]

== 2 of 2 ==
Date: Wed, Dec 16 2009 10:51 pm
From: Patrick Connolly

On Thu, 17-Dec-2009 at 03:17PM +1000, Zoe van Havre wrote:

[...]

|> The problem is that I have been tracking one factor in particular
|> ('branch', values 2 or 3) and once the final merge occurs, the
|> second level of this factor seems to disappear in the last dataset,
|> even though it was present before. See code & output below:

|>
|> > dim(tma)

You didn't tell us that one. What size is it?

|> > names(tma)
|> [1] "Code" "marker" "cell" "tumourA" "tumourEXP" "int" "stain" "tumourPERC" "branch"
|> > levels(tma$tumourA)
|> [1] "DCIS" "LN Metastasis" "Normal" "Primary Invasive Carcinoma"
|> #split into cancer and normal tissue
|> > tma1<-subset(tma, tumourA=="Primary Invasive Carcinoma")
|> > tma2<-subset(tma, tumourA=="LN Metastasis")
|> > tmaN<-subset(tma, tumourA=="Normal")
|>

[...]

|> 2 3
|> 91 51
|> > table(tma1.1$branch.x)
|>
|> 2 3
|> 1806 633
|> > table(tma2.1$branch.x)
|>
|> 3
|> 625
|>
|>
|> Please, can someone tell me what's going on?

I suspect you'd have a lot of NAs in there. Try this:
sapply(tma, function(x)
sum(is.na(x)))

If that doesn't tell you something interesting, try with the subsets.
Or maybe when you use table(), try the exclude=NULL argument.

HTH

--
~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.
___ Patrick Connolly
{~._.~} Great minds discuss ideas
_( Y )_ Average minds discuss events
(:_~*~_:) Small minds discuss people
(_)-(_) ..... Eleanor Roosevelt

~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.

==============================================================================
TOPIC: change the scale of grid lines in a plot
http://groups.google.com/group/r-help-archive/t/ed2627a0ec1bf29c?hl=en
==============================================================================

== 1 of 1 ==
Date: Wed, Dec 16 2009 9:35 pm
From: Jim Lemon

On 12/17/2009 02:22 PM, Roslina Zakaria wrote:
> Hi,
>
> I would like to change a scale of the grid lines in a plot. I can see the default is 0.2, so how do I change to 0.1?
>
>
Hi Roslina,
This depends very much on the plot you are using. "grid", which will add
grid lines to a base graphics plot, wants to know how many cells to
draw, not how far apart the lines should be in user (?) units. What
graphing package are you using?

Jim

== 1 of 2 ==
Date: Wed, Dec 16 2009 10:25 pm
From: Maithili Shiva

Dear R helpers

I am having following data

Name           Numbers
A                  25
B                   3
C                  13
A                   5
B                   7
C                   0
A                   2
B                 10
C                  9

CONDITIONS

If Name is A, min_val = 1.05, max_val = 1.30
If Name is B, min_val = 1.30, max_val = 1.60
If Name is C, min_val = 1.60, max_val = 1.99

TASK

To generate the Uniform random nos for each of these Names (Equal to the corresponding no. e.g. the 5th Name is B, so I need to generate 7 uniform random numbers in the range (1.30 - 1.99). Also I need to arrange these random numbers one by one in a single csv file i.e. say 25 random numbers in teh range (1.05-1.30) followed by 3 random numbers in the range (1.30-1.60) and so on.

# ___________________________________________________

Here is the R code I have tried

ONS <- read.table(textConnection("name number
A11    12
A12    17
A13     0
A11    11
A12      6
A13      0
A11      8
A12      4
A13      3"), header = TRUE)

X = as.character(ONS$name)
Y = ONS$number

Z = NULL

for (i in 1:length(X))
   {
   if(X[i] == 'A11')
   {
   min_val = 1.05
   max_val = 1.30
   Z = runif(Y[i], min_val, max_val)
   }
   else
   {
   if(X[i] == 'A12')
   {
   min_val = 1.30
   max_val = 1.60
   Z = runif(Y[i], min_val, max_val)
   }
   else
   {
   if(X[i] == 'A13')
   {
   min_val = 1.60
   max_val = 1.99
   Z = runif(Y[i], min_val, max_val)
   }
   }
   }
   }

# End of Code

## _____________________________________________________________________

PROBLEM

I need to get 61 random numbers which is total of all the numbers (1st 12 fo A, 3 random numbers for B, 13 for C, 5 again fo A and so on). The result whcih I got is

> Z
[1] 1.740443 1.761758 1.797222

which is pertaining to the last name C where 3 random numbers are generated i.e. Z instaed of getting added, is overwritten.

Please help me to rectify my code so that in the end I will get 61 random numbers as desired i.e. 12 for A in the range (1.05 - 1.30), 3 for B in the range (1.30 - 1.60), 13 for C in the range (1.60-1.99), again 5 for A in the range (1.05 - 1.30).

Thanking in advance. I also sincerely apologize for writing such a long mail, as I wanted to be clear as possible in my communication.

Regards

Maithili

The INTERNET now has a personality. YOURS! See your Yahoo! Homepage.
[[alternative HTML version deleted]]

== 2 of 2 ==
Date: Wed, Dec 16 2009 11:23 pm
From: "Daniel Malter"

Hi, does this do what you want to do? I take the first six lines of the data
you provided. Note that "letters" corresponds to your "Name," and "number"
corresponds to your "Numbers" variable.

letters=rep(c("A","B","C"),2)
letters

number=c(25,3,13,5,7,0)
number

letters2=rep(letters,number)
letters2

interval.min=ifelse(letters2=="A",1.05,ifelse(letters2=="B",1.3,1.6))
interval.max=ifelse(letters2=="A",1.3,ifelse(letters2=="B",1.6,1.99))

rand.nums=runif(letters2,min=interval.min,max=interval.max)

data.frame(letters2,rand.nums)

HTH,
Daniel

The INTERNET now has a personality. YOURS! See your Yahoo! Homepage.
[[alternative HTML version deleted]]

== 1 of 1 ==
Date: Wed, Dec 16 2009 10:35 pm
From: Maithili Shiva

Dear R helpers, please ignore my earlier mail. Here is the corrected mail. Please forgive me for the lapses on my part. Extremely sorry.

Here is the corrected mail.

Dear R helpers

I am having following data

Name           Numbers
A11                  12
A12                  17
A13                   0
A11                  11
A12                   6
A13                   0
A11                   8
A12                   4
A13                   3

CONDITIONS

If Name is A11, min_val = 1.05, max_val = 1.30
If Name is A12, min_val = 1.30, max_val = 1.60
If Name is A13, min_val = 1.60, max_val = 1.99

TASK

To generate the Uniform random nos for each of these Names (Equal to the corresponding no. e.g. the 5th Name is A12, so I need to generate 6 uniform random numbers in the range (1.30 - 1.99). Also I need to arrange these random numbers one by one in a single csv file i.e. say 12 random numbers in teh range (1.05-1.30) followed by 17 random numbers in the range (1.30-1.60) and so on.

# ___________________________________________________

Here is the R code I have tried

ONS <- read.table(textConnection("name number
A11    12
A12    17
A13     0
A11    11
A12      6
A13      0
A11      8
A12      4
A13      3"), header = TRUE)

X = as.character(ONS$name)
Y = ONS$number

Z = NULL

for (i in 1:length(X))
   {
   if(X[i] == 'A11')
   {
   min_val = 1.05
   max_val = 1.30
   Z = runif(Y[i], min_val, max_val)
   }
   else
   {
   if(X[i] == 'A12')
   {
   min_val = 1.30
   max_val = 1.60
   Z = runif(Y[i], min_val, max_val)
   }
   else
   {
   if(X[i] == 'A13')
   {
   min_val = 1.60
   max_val = 1.99
   Z = runif(Y[i], min_val, max_val)
   }
   }
   }
   }

# End of Code

## _____________________________________________________________________

PROBLEM

I need to get 61 random numbers which is total of all the numbers (1st 12 fo A, 3 random numbers for B, 13 for C, 5 again fo A and so on). The result whcih I got is

> Z
[1] 1.740443 1.761758 1.797222

which is pertaining to the last name C where 3 random numbers are generated i.e. Z instaed of getting added, is overwritten.

Please help me to rectify my code so that in the end I will get 61 random numbers as desired i.e. 12 for A in the range (1.05 - 1.30), 3 for B in the range (1.30 - 1.60), 13 for C in the range (1.60-1.99), again 5 for A in the range (1.05 - 1.30).

Thanking in advance. I also sincerely apologize for writing such a long mail, as I wanted to be clear as possible in my communication.

Regards

Maithili

The INTERNET now has a personality. YOURS! See your Yahoo! Homepage.
[[alternative HTML version deleted]]

==============================================================================

You received this message because you are subscribed to the Google Groups "R-help-archive"
group.

To post to this group, send email to r-help@r-project.org or visit http://groups.google.com/group/r-help-archive?hl=en

To unsubscribe from this group, send email to r-help-archive+unsubscribe@googlegroups.com

To change the way you get mail from this group, visit:
http://groups.google.com/group/r-help-archive/subscribe?hl=en

To report abuse, send email explaining the problem to abuse@googlegroups.com

==============================================================================
Google Groups: http://groups.google.com/?hl=en

r-help-archive - 25 new messages in 14 topics - digest

2009-12-16T21:10:00.001-08:00

R-help-archive
http://groups.google.com/group/r-help-archive?hl=en

r-help-archive@googlegroups.com

Today's topics:

==============================================================================
TOPIC: fortune-like FAQ search
http://groups.google.com/group/r-help-archive/t/b08c4fffe3511268?hl=en
==============================================================================

== 1 of 1 ==
Date: Wed, Dec 16 2009 11:11 am
From: baptiste auguie

Dear list,

I'm not so familiar with the internals of the fortunes package, but I
really like the interface. I was wondering if someone had implemented
a similar functionality to parse the entries of the R FAQ <
http://cran.r-project.org/doc/FAQ/R-FAQ.html >. Say, if I was to
answer a question "Why does my lattice graphic fail to produce an
output in this function?", I might try:

library(rfaq)
faq("lattice")
## or if I were a number kind of person
faq(722)

The database could be split into "windows only", etc.

Should this approach prove useful, one could even imagine using this
rfaq database as the original source to produce the current html page.
The infrastructure might also be of interest for other packages which
could provide their own database.

Just a thought anyway,

Best regards,

baptiste

==============================================================================
TOPIC: number of observations used in cor when use="pairwise.obs"
http://groups.google.com/group/r-help-archive/t/b5620e5c0a352564?hl=en
==============================================================================

== 1 of 3 ==
Date: Wed, Dec 16 2009 11:47 am
From: Jean-Christophe Domenge

Dear R gurus,
to compute the correlation matrix of "n" variables with "n_obs" observations
each,
possibly including NA, I use cor(M, use="pairwise.obs")
where m is a "n" x "nobs" matrix.

Now I want to know the number of observations actually used in this
computation,
namely for each pair of columns in M, say pair (i,j), I want to compute
sum( !is.na(M[,i]) & !is.na(M[,j]) ).
I can think of several ways of constructing all pairs (i,j) with i<j, then
computing the sum
above, but for n=2000 and nobs=700 the computation time is prohibitive.

Is there an efficient way to solve this?

Many thanks,

[[alternative HTML version deleted]]

== 2 of 3 ==
Date: Wed, Dec 16 2009 11:55 am
From: "William Dunlap"

> -----Original Message-----
> From: r-help-bounces@r-project.org
> [mailto:r-help-bounces@r-project.org] On Behalf Of
> Jean-Christophe Domenge
> Sent: Wednesday, December 16, 2009 11:48 AM
> To: r-help
> Subject: [R] number of observations used in cor when
> use="pairwise.obs"
>
> Dear R gurus,
> to compute the correlation matrix of "n" variables with
> "n_obs" observations
> each,
> possibly including NA, I use cor(M, use="pairwise.obs")
> where m is a "n" x "nobs" matrix.
>
> Now I want to know the number of observations actually used in this
> computation,
> namely for each pair of columns in M, say pair (i,j), I want
> to compute
> sum( !is.na(M[,i]) & !is.na(M[,j]) ).
> I can think of several ways of constructing all pairs (i,j)
> with i<j, then
> computing the sum
> above, but for n=2000 and nobs=700 the computation time is
> prohibitive.

Try
crossprod(!is.na(M))

Bill Dunlap
Spotfire, TIBCO Software
wdunlap tibco.com

>
> Is there an efficient way to solve this?
>
> Many thanks,
>
> jc
>
> [[alternative HTML version deleted]]
>
> ______________________________________________
> R-help@r-project.org mailing list
> https://stat.ethz.ch/mailman/listinfo/r-help
> PLEASE do read the posting guide
> http://www.R-project.org/posting-guide.html
> and provide commented, minimal, self-contained, reproducible code.
>

== 3 of 3 ==
Date: Wed, Dec 16 2009 11:59 am
From: Jean-Christophe Domenge

great, thanks!

On Wed, Dec 16, 2009 at 20:55, William Dunlap <wdunlap@tibco.com> wrote:

> > -----Original Message-----
> > From: r-help-bounces@r-project.org
> > [mailto:r-help-bounces@r-project.org] On Behalf Of
> > Jean-Christophe Domenge
> > Sent: Wednesday, December 16, 2009 11:48 AM
> > To: r-help
> > Subject: [R] number of observations used in cor when
> > use="pairwise.obs"
> >
> > Dear R gurus,
> > to compute the correlation matrix of "n" variables with
> > "n_obs" observations
> > each,
> > possibly including NA, I use cor(M, use="pairwise.obs")
> > where m is a "n" x "nobs" matrix.
> >
> > Now I want to know the number of observations actually used in this
> > computation,
> > namely for each pair of columns in M, say pair (i,j), I want
> > to compute
> > sum( !is.na(M[,i]) & !is.na(M[,j]) ).
> > I can think of several ways of constructing all pairs (i,j)
> > with i<j, then
> > computing the sum
> > above, but for n=2000 and nobs=700 the computation time is
> > prohibitive.
>
> Try
> crossprod(!is.na(M))
>
> Bill Dunlap
> Spotfire, TIBCO Software
> wdunlap tibco.com
>
> >
> > Is there an efficient way to solve this?
> >
> > Many thanks,
> >
> > jc
> >
> > [[alternative HTML version deleted]]
> >
> > ______________________________________________
> > R-help@r-project.org mailing list
> > https://stat.ethz.ch/mailman/listinfo/r-help
> > PLEASE do read the posting guide
> > http://www.R-project.org/posting-guide.html
> > and provide commented, minimal, self-contained, reproducible code.
> >
>

[[alternative HTML version deleted]]

==============================================================================
TOPIC: write.csv and header
http://groups.google.com/group/r-help-archive/t/d6e8e024ed1218ce?hl=en
==============================================================================

== 1 of 4 ==
Date: Wed, Dec 16 2009 11:49 am
From: Patrick Connolly

On Tue, 15-Dec-2009 at 01:55PM +0100, Gustaf Rydevik wrote:

|> Hi,
|>
|> ?write.table and the argument "append" should be of help.
|> example:
|>
|> > sink("test.csv")
|> > cat("-----------------------------")
|> > cat("\n")
|> > cat("This is \n a test of header")
|> > cat("\n")
|> > cat("-----------------------------")
|> > cat("\n")
|> > sink()
|> >

Or, instead of the use of sink() and cat(), one could use the write()
function, also using the append = TRUE argument.

|> write.table(matrix(rnorm(100),nrow=10),file="test.csv",append=TRUE,sep=",")
|>
|>
|> regards,
|> Gustaf
|>
|>
|> --
|> Gustaf Rydevik, M.Sci.
|> tel: +46(0)703 051 451
|> address:Essingetorget 40,112 66 Stockholm, SE
|> skype:gustaf_rydevik
|>
|> [[alternative HTML version deleted]]
|>
|> ______________________________________________
|> R-help@r-project.org mailing list
|> https://stat.ethz.ch/mailman/listinfo/r-help
|> PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
|> and provide commented, minimal, self-contained, reproducible code.

== 2 of 4 ==
Date: Wed, Dec 16 2009 12:04 pm
From:

I have a small request regarding this "append" feature. As it is now if the data is appended to the file so is the header. I would like to have the header only entered once and appends just append the data.

Doable?

Kevin

---- Patrick Connolly <p_connolly@slingshot.co.nz> wrote:
> On Tue, 15-Dec-2009 at 01:55PM +0100, Gustaf Rydevik wrote:
>
>
> |> Hi,
> |>
> |> ?write.table and the argument "append" should be of help.
> |> example:
> |>
> |> > sink("test.csv")
> |> > cat("-----------------------------")
> |> > cat("\n")
> |> > cat("This is \n a test of header")
> |> > cat("\n")
> |> > cat("-----------------------------")
> |> > cat("\n")
> |> > sink()
> |> >
>
> Or, instead of the use of sink() and cat(), one could use the write()
> function, also using the append = TRUE argument.
>
>
> |> write.table(matrix(rnorm(100),nrow=10),file="test.csv",append=TRUE,sep=",")
> |>
> |>
> |> regards,
> |> Gustaf
> |>
> |>
> |> --
> |> Gustaf Rydevik, M.Sci.
> |> tel: +46(0)703 051 451
> |> address:Essingetorget 40,112 66 Stockholm, SE
> |> skype:gustaf_rydevik
> |>
> |> [[alternative HTML version deleted]]
> |>
> |> ______________________________________________
> |> R-help@r-project.org mailing list
> |> https://stat.ethz.ch/mailman/listinfo/r-help
> |> PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
> |> and provide commented, minimal, self-contained, reproducible code.
>
> --
> ~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.
> ___ Patrick Connolly
> {~._.~} Great minds discuss ideas
> _( Y )_ Average minds discuss events
> (:_~*~_:) Small minds discuss people
> (_)-(_) ..... Eleanor Roosevelt
>
> ~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.
>
> ______________________________________________
> R-help@r-project.org mailing list
> https://stat.ethz.ch/mailman/listinfo/r-help
> PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
> and provide commented, minimal, self-contained, reproducible code.

== 3 of 4 ==
Date: Wed, Dec 16 2009 12:05 pm
From:

Doable?

Kevin

== 4 of 4 ==
Date: Wed, Dec 16 2009 2:55 pm
From: John Kane

Just write the headers once and append as many times as needed?

You can append as many datasets as you want.

--- On Wed, 12/16/09, rkevinburton@charter.net <rkevinburton@charter.net> wrote:

> From: rkevinburton@charter.net <rkevinburton@charter.net>
> Subject: Re: [R] write.csv and header
> To: "Gustaf Rydevik" <gustaf.rydevik@gmail.com>, "Patrick Connolly" <p_connolly@slingshot.co.nz>
> Cc: r-help@r-project.org
> Received: Wednesday, December 16, 2009, 3:05 PM
> I have a small request regarding this
> "append" feature. As it is now if the data is appended to
> the file so is the header. I would like to have the header
> only entered once and appends just append the data.
>
> Doable?
>
> Kevin
>
> ---- Patrick Connolly <p_connolly@slingshot.co.nz>
> wrote:
> > On Tue, 15-Dec-2009 at 01:55PM +0100, Gustaf Rydevik
> wrote:
> >
> >
> > |> Hi,
> > |>
> > |> ?write.table and the argument "append" should be
> of help.
> > |> example:
> > |>
> > |> > sink("test.csv")
> > |> > cat("-----------------------------")
> > |> > cat("\n")
> > |> > cat("This is \n a test of header")
> > |> > cat("\n")
> > |> > cat("-----------------------------")
> > |> > cat("\n")
> > |> > sink()
> > |> >
> >
> > Or, instead of the use of sink() and cat(), one could
> use the write()
> > function, also using the append = TRUE argument.
> >
> >
> > |>
> write.table(matrix(rnorm(100),nrow=10),file="test.csv",append=TRUE,sep=",")
> > |>
> > |>
> > |> regards,
> > |> Gustaf
> > |>
> > |>
> > |> --
> > |> Gustaf Rydevik, M.Sci.
> > |> tel: +46(0)703 051 451
> > |> address:Essingetorget 40,112 66 Stockholm, SE
> > |> skype:gustaf_rydevik
> > |>
> > |>     [[alternative HTML version
> deleted]]
> > |>
> > |> ______________________________________________
> > |> R-help@r-project.org
> mailing list
> > |> https://stat.ethz.ch/mailman/listinfo/r-help
> > |> PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
> > |> and provide commented, minimal, self-contained,
> reproducible code.
> >
> > --
> >
> ~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.
> > ___ Patrick
> Connolly
> > {~._.~}
>    Great minds discuss
> ideas
> > _( Y )_
>    Average minds discuss events
> > (:_~*~_:)
> Small minds discuss people
> > (_)-(_)
>
> ..... Eleanor Roosevelt
> >
> >
> ~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.
> >
> > ______________________________________________
> > R-help@r-project.org
> mailing list
> > https://stat.ethz.ch/mailman/listinfo/r-help
> > PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
> > and provide commented, minimal, self-contained,
> reproducible code.
>
> ______________________________________________
> R-help@r-project.org
> mailing list
> https://stat.ethz.ch/mailman/listinfo/r-help
> PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
> and provide commented, minimal, self-contained,
> reproducible code.
>

__________________________________________________________________
Be smarter than spam. See how smart SpamGuard is at giving ju
______________________________________________
R-help@r-project.org mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.

==============================================================================
TOPIC: memory problem on Suse
http://groups.google.com/group/r-help-archive/t/f9d0d9e1a23ce4d2?hl=en
==============================================================================

== 1 of 1 ==
Date: Wed, Dec 16 2009 5:02 am
From: Ambrosi Alessandro

Dear all and dear Marc, it seems you hit the target.
I checked as you suggested, and... it is a 32 bit version!
Now I'm fixing it. Thank you very much.
Alessandro

________________________________________
From: Marc Schwartz [marc_schwartz@me.com]
Sent: 11 December 2009 17:02
To: Ambrosi Alessandro
Cc: r-help@r-project.org
Subject: Re: [R] memory problem on Suse

On Dec 11, 2009, at 6:24 AM, Ambrosi Alessandro wrote:

>
> Dear all, I am meeting some problems with memory allocation. I know
> it is an old issue, I'm sorry.
> I looked for a solution in the FAQs and manuals, mails, but without
> finding the working answer.
> I really hope you can help me.
> For instance, if I try to read micorarray data I get:
>
>> mab=ReadAffy(cdfname="hgu133plus2cdf")
> Error: cannot allocate vector of size 858.0 Mb
>>
>
> I get similar errors with smaller objects, smaller data sets or
> other procedures
> ("Error: cannot allocate vector of size 123.0 Mb").
> I'm running R with Suse 11.1 Linux OS, on two Xeon processors (8
> cores), 32 GB RAM.
> I suppose I have enough resources to manage these objects and data
> files....
>
> Any suggestions or hints will be really appreciated!
> Many thanks in advance.
> Alessandro

Well, you are running into a situation where there is not a contiguous
chunk of RAM available in the sizes referenced, for allocation to the
vector.

Presuming that you are running a 64 bit version of SUSE (what does
'uname -a' show in a system console), you should also check to be sure
that you are also running a 64 bit version of R. What does:

.Machine$sizeof.pointer

show?

If it returns 4, then you are running a 32 bit version of R, which
cannot take advantage of your 64 bit platform. You should install a 64
bit version of R.

HTH,

Marc Schwartz
______________________________________________
R-help@r-project.org mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.

==============================================================================
TOPIC: Rgraphviz install
http://groups.google.com/group/r-help-archive/t/22b7757367081dd5?hl=en
==============================================================================

== 1 of 1 ==
Date: Wed, Dec 16 2009 8:36 am
From: Costa

Hi,

I installed R.10.1 and Graphviz 2.26 - which doesnt contain the
libcdt-4.dll, so I get the same error message 'This application has failed
to start because libcdt-4.dll was not found' when loading in R
> library (Rgraphviz)

I cannot find an installation file for Graphviz 2.20 to use under Windows7
(which was indicated previously to solve this problem).
Can anybody tell me where I could find this installation file?

Thanks ...

Uwe Ligges-3 wrote:
>
>
>
> June Wong wrote:
>> Hi
>>
>> I've tried several times and followed what others suggest in installing
>> Rgraphviz
>> (https://stat.ethz.ch/pipermail/bioconductor/2008-June/022838.html)
>> But I could not load library(Rgraphviz).
>> There's a message that 'This application has failed to start because
>> libcdt-4.dll was not found'
>>
>> Could anyone suggest me how to make it work?
>
>
> Yes, install the "latest" graphviz from the installer graphviz-2.20.3.msi.
>
> Note that it install under a directory called .....2.21
> So no idea which version it is (and hard to find out).
>
> Best,
> Uwe Ligges
>
>
>
>
>> june
>> _________________________________________________________________
>>
>>
>> ore_012009
>> [[alternative HTML version deleted]]
>>
>> ______________________________________________
>> R-help@r-project.org mailing list
>> https://stat.ethz.ch/mailman/listinfo/r-help
>> PLEASE do read the posting guide
>> http://www.R-project.org/posting-guide.html
>> and provide commented, minimal, self-contained, reproducible code.
>
> ______________________________________________
> R-help@r-project.org mailing list
> https://stat.ethz.ch/mailman/listinfo/r-help
> PLEASE do read the posting guide
> http://www.R-project.org/posting-guide.html
> and provide commented, minimal, self-contained, reproducible code.
>
>

--
View this message in context: http://n4.nabble.com/Rgraphviz-install-tp878526p965352.html
Sent from the R help mailing list archive at Nabble.com.

==============================================================================
TOPIC: regularised incomplete beta function
http://groups.google.com/group/r-help-archive/t/978801eb8500cac6?hl=en
==============================================================================

== 1 of 3 ==
Date: Wed, Dec 16 2009 3:31 am
From: "Xin Shi"

Dear:

I am trying to work out the regularised incomplete beta function in R. I searched google
on this and found UCS library. But I can not find this in R packages. Does anyone have use
this before or how to insert UCS in R?

Many Thanks!

Xin

Xin
-------
Dr.Xin Shi
Senior Lecturer in Statistics
Manchester Metropolitan University Business School
Aytoun Building
Aytoun Street
Manchester
M1 3GH
United Kingdom

Telephone: +44 (0)161 247 3841
Fax: +44(0)161 247 6916
E-mail: X.Shi@mmu.ac.uk

Before acting on this email or opening any attachments you should read the Manchester
Metropolitan University's email disclaimer
available on its website:
http://www.mmu.ac.uk/emaildisclaimer

[[alternative HTML version deleted]]

== 2 of 3 ==
Date: Wed, Dec 16 2009 12:40 pm
From: "Ravi Varadhan"

?pbeta

----------------------------------------------------------------------------
-------

Ravi Varadhan, Ph.D.

Assistant Professor, The Center on Aging and Health

Division of Geriatric Medicine and Gerontology

Johns Hopkins University

Ph: (410) 502-2619

Fax: (410) 614-9625

Email: rvaradhan@jhmi.edu

Webpage:
http://www.jhsph.edu/agingandhealth/People/Faculty_personal_pages/Varadhan.h
tml

----------------------------------------------------------------------------
--------

-----Original Message-----
From: r-help-bounces@r-project.org [mailto:r-help-bounces@r-project.org] On
Behalf Of Xin Shi
Sent: Wednesday, December 16, 2009 6:32 AM
To: r-help@stat.math.ethz.ch
Subject: [R] regularised incomplete beta function

Dear:

I am trying to work out the regularised incomplete beta function in R. I
searched google
on this and found UCS library. But I can not find this in R packages. Does
anyone have use
this before or how to insert UCS in R?

Many Thanks!

Xin

Xin
-------
Dr.Xin Shi
Senior Lecturer in Statistics
Manchester Metropolitan University Business School
Aytoun Building
Aytoun Street
Manchester
M1 3GH
United Kingdom

Telephone: +44 (0)161 247 3841
Fax: +44(0)161 247 6916
E-mail: X.Shi@mmu.ac.uk

Before acting on this email or opening any attachments you should read the
Manchester
Metropolitan University's email disclaimer
available on its website:
http://www.mmu.ac.uk/emaildisclaimer

[[alternative HTML version deleted]]

== 3 of 3 ==
Date: Wed, Dec 16 2009 2:42 pm
From: Stefan Evert

On 16 Dec 2009, at 21:40, Ravi Varadhan wrote:

> ?pbeta

And the zipfR package wraps these in terms of the usual terminology
for incomplete/regularised upper/lower Beta functions (see ?Rbeta
there), for people like me who can't get their head around the
equivalence between the Beta distribution and the various versions of
the Beta function. :-)

Cheers,
Stefan

==============================================================================
TOPIC: WinBUGS - R2WinBUGS problem
http://groups.google.com/group/r-help-archive/t/b209d64de84032ff?hl=en
==============================================================================

== 1 of 1 ==
Date: Wed, Dec 16 2009 8:58 am
From: Sotiris Adamakis

Appologies for cross-posting

Dear R users,

I am using R2WinBUGS to call WinBUGS from R. After loading data, model, and initial values I call this command

res <- bugs(data = dfile, inits = list(ifile), parameters.to.save = c("beta"), model.file = mfile, working.directory = tdir, n.thin = nthin, n.chains = 1, n.iter = niterations*nthin, n.burnin = 0, DIC = F, debug = T, save.history=F)

which opens WinBUGS and executes the following log file:

display(log)

check(C:/Documents and Settings/sbadamakis/Desktop/model_4/model_4.txt)

model is syntactically correct

data(C:/Documents and Settings/sbadamakis/Desktop/model_4/output1/data.txt)

data loaded

compile(1)

model compiled

inits(1,C:/Documents and Settings/sbadamakis/Desktop/model_4/output1/inits1.txt)

model is initialized

gen.inits()

command #Bugs:gen.inits cannot be executed (is greyed out)

thin.updater(1)

update(0)

set(beta)

update(20)

coda(*,C:/Documents and Settings/sbadamakis/Desktop/model_4/output1/coda)

stats(*)

Node statistics

node mean sd MC error 2.5% median 97.5% start sample

beta[1] 1.588 0.009812 0.003936 1.579 1.584 1.607 1 20

beta[2] -0.04272 0.00263 0.001137 -0.04592 -0.0433 -0.03688 1 20

beta[3] -0.07878 8.531E-4 2.829E-4 -0.07962 -0.07833 -0.07699 1 20

beta[4] -0.3035 0.01016 0.004509 -0.3194 -0.3028 -0.2841 1 20

beta[5] 1.241 0.00854 0.002739 1.228 1.242 1.258 1 20

beta[6] -4.155 0.01242 0.005405 -4.174 -4.152 -4.132 1 20

beta[7] 0.08322 0.001485 6.379E-4 0.08201 0.0824 0.08608 1 20

beta[8] -3.001 0.009335 0.003318 -3.013 -3.004 -2.981 1 20

beta[9] 0.07953 7.547E-4 3.198E-4 0.07777 0.07932 0.08035 1 20

save(C:/Documents and Settings/sbadamakis/Desktop/model_4/output1/log.odc)

save(C:/Documents and Settings/sbadamakis/Desktop/model_4/output1/log.txt)

So far so good. After the end of the last line of the log file, WinBUGS should close and the bugs function in R should return the Markov chain values of the "beta" parameters. The problem is that it doesn't and WinBUGS stays infinitely open. This is a big problem as I have written a code in R and I want to call the bugs function several times. Do you know if there's an option in the bugs function in R to FORCE WinBUGS to close?

Furthermore, some other times (the same code!!!) produces the following TRAP message

undefined real result

GraphStack.Node.Value [00000C9AH]
.i INTEGER 6
.j INTEGER 2
.k INTEGER 1
.node GraphStack.Node [0138C5D0H]
.op INTEGER 6
.p GraphNodes.Node [01544D30H]
.top INTEGER 0
.work ARRAY 10 OF REAL
[0] REAL inf
[1] REAL inf
[2] REAL inf
[3] REAL 1.30212336033939
[4] REAL 4.534462964264759
[5] REAL 1.769649991139459E+167
[6] REAL 1.792592453177908E-307
[7] REAL 9.94060079432588E-321
[8] REAL 5.474495960374562E+67
[9] REAL 5.257527858867595E-308
MonitorsSamples.StdMonitor.Update [0000023DH]
.blocks POINTER NIL
.i INTEGER 1655837286
.index INTEGER 7
.monitor MonitorsSamples.StdMonitor [02155B50H]
.numBlocks INTEGER 0
.value REAL 1.0
SamplesMonitors.Monitor.Update [0000091BH]
.chain INTEGER 0
.i INTEGER 4480
.monitor SamplesMonitors.Monitor [01C0A4E0H]
.size INTEGER 6780
SamplesIndex.Monitor.Update [0000030CH]
.chain INTEGER 0
.cursor SamplesIndex.List [01AB2E30H]
.monitor SamplesIndex.Monitor [01C59510H]
.sampleMonitor SamplesMonitors.Monitor [01C0A4E0H]
UpdaterActions.UpdateMonitors [0000011CH]
.chain INTEGER 0
.monitor UpdaterActions.Monitor [01C59510H]
BugsCmds.Action.Step [00000BCAH]
.a BugsCmds.Action [011D40C0H]
.chain INTEGER 0
.i INTEGER 8
.j INTEGER 1000
.numChains INTEGER 1
BugsCmds.Action.Do [00000C58H]
.a BugsCmds.Action [011D40C0H]
.elapsedTime LONGINT 62EC21F162EB0384H
.profileList ARRAY 2024 OF CHAR "" ...
.res INTEGER 1635734111
.s ARRAY 120 OF CHAR 3X ...
Services.Exec [00000136H]
.a Services.Action [011D40C0H]
.t POINTER [62EB0384H]
Services.IterateOverActions [000002F4H]
.p Services.Action [011D40C0H]
.t POINTER NIL
.time LONGINT 300843187
Services.StdHook.Step [0000034DH]
.h Services.StdHook [0101E380H]
HostWindows.Idle [00004A86H]
.focus BOOLEAN FALSE
.tick Controllers.TickMsg Fields
.w HostWindows.Window NIL
HostMenus.TimerTick [00003422H]
.lParam INTEGER 0
.ops Controllers.PollOpsMsg Fields
.wParam INTEGER 1
.wnd INTEGER 1770006
Kernel.Try [00003A61H]
.a INTEGER 1770006
.b INTEGER 1
.c INTEGER 0
.h PROCEDURE HostMenus.TimerTick
HostMenus.ApplWinHandler [00003841H]
.Proc PROCEDURE NIL
.hit BOOLEAN FALSE
.lParam INTEGER 0
.message INTEGER 275
.res INTEGER 4243644
.s ARRAY 256 OF SHORTCHAR "¯B", 1X ...
.w INTEGER 2147348480
.wParam INTEGER 1
.wnd INTEGER 1770006
<system> (pc=7E418733H, fp=0022FB84H)
<system> (pc=7E418815H, fp=0022FBECH)
<system> (pc=7E4189CCH, fp=0022FC4CH)
<system> (pc=7E4196C6H, fp=0022FC5CH)
HostMenus.Loop [00003BDEH]
.done BOOLEAN FALSE
.f SET {0..5}
.n INTEGER 0
.res INTEGER 0
.w HostWindows.Window NIL
Kernel.Start [00002B8CH]
.code PROCEDURE HostMenus.Loop

Is there anyone who can explain what does this mean and how to understand trap messages? For me, it seems that the problem lies here:

.work ARRAY 10 OF REAL
[0] REAL inf
[1] REAL inf
[2] REAL inf
[3] REAL 1.30212336033939
[4] REAL 4.534462964264759
[5] REAL 1.769649991139459E+167
[6] REAL 1.792592453177908E-307
[7] REAL 9.94060079432588E-321
[8] REAL 5.474495960374562E+67
[9] REAL 5.257527858867595E-308

where it says that it produces some inf parameter values. I do not understand why this happens as the program was working fine for 5*10^5 iterations. I tried to change the Seed values but there was no use.

A bit of background:
1) I am running multilevel models for small area estimation and the above problems seem to appear only with the spatial random effects model. The spatial+non-spatial random effects model seems to behave properly without any problems.

2) As mentioned above the program was working fine for more than 500000 iterations. This means that I got no trap messages and I called the bugs function for more than 10 times without any problems.

It's so frustrating...

Kind regards,
Sotiris

Dr. Sotiris Adamakis
Senior Statistician - Ipsos MORI

T +44 20 7347 3828
s<mailto:ssotiris.adamakis@ipsos.com>otiris.adamakis<mailto:ssotiris.adamakis@ipsos.com>@ipsos.com<mailto:ssotiris.adamakis@ipsos.com>
www.ipsos-mori.com<http://www.ipsos-mori.com/>

79-81 Borough Road, London, SE1 1FY

________________________________
================================================================================================
This e-mail and all attachments it may contain is confidential and intended solely for the use of the individual to whom it is addressed. Any views or opinions presented are solely those of the author and do not necessarily represent those of Ipsos MORI and its associated companies. If you are not the intended recipient, be advised that you have received this e-mail in error and that any use, dissemination, printing, forwarding or copying of this e-mail is strictly prohibited. Please contact the sender if you have received this e-mail in error.

Market & Opinion Research International Ltd (trading as Ipsos MORI), Registered in England and Wales No. 948470, 79-81 Borough Road, London SE1 1FY, United Kingdom. VAT reg no. 443932151. Email: ukinfo@ipsos.com

================================================================================================

[[alternative HTML version deleted]]

==============================================================================
TOPIC: henon map
http://groups.google.com/group/r-help-archive/t/9440be1de459f881?hl=en
==============================================================================

== 1 of 1 ==
Date: Wed, Dec 16 2009 12:43 pm
From: x y

Hi, I would plot a Henon map:

> plot(henon(start=rnorm(2), a=1.4, b=0.3, n.sample=2000, n.transient=10),pch=".",cex=0.1)

but give me some errors:

Error in henon(start = rnorm(2), a = 1.4, b = 0.3, n.sample = 2000, n.transient = 10) :
could not find function "checkVectorType"
Error in plot(henon(start = rnorm(2), a = 1.4, b = 0.3, n.sample = 2000, :
error in evaluating the argument 'x' in selecting a method for function 'plot'

I've loaded "fractal" package. An other fractal, "lorenz", work well.
I use linux mandriva 2010.0
Any suggestions?
Thanks
:)

_________________________________________________________________

[[alternative HTML version deleted]]

== 1 of 1 ==
Date: Wed, Dec 16 2009 1:13 pm
From: Peng Yu

Currently, I load the RData file then ls() and str(). But loading the file
takes too long if the file is big. Most of the time, I only interested what
the variables are in the the file and the attributes of the variables (like
if it is a data.frame, matrix, what are the colnames/rownames, etc.)

I'm wondering if there is any facility in R to help me avoid loading the
whole file.

[[alternative HTML version deleted]]

==============================================================================
TOPIC: Odp: Creating Dummy Variables in R
http://groups.google.com/group/r-help-archive/t/c9746670ecbbf081?hl=en
==============================================================================

== 1 of 4 ==
Date: Wed, Dec 16 2009 2:14 pm
From: "whitaker m. (mw1006)"

I have a much larger dataset than in my original email (attached - price dependent upon weight, Clarity (different levels IF-SI2), colour(levels D-L) and Cut (ideal-fair), and tried the regression command:

>diamond.lm<-lm(price~weight+IF+VVS1+VVS2+VS1+VS2+SI1+SI2+I1+I2+D+E+F+G+H+I+J+K+L+ideal+excellent+very.good+good+fair, data="Diamonds2.txt")

Error in eval(predvars, data, env) : invalid 'envir' argument

Which lead to the error message below the command. I have tried searching for this, and assumed this was down to having categrocial variables within the data, is this a correct assumption or am i doing something else wrong? Apologies if this is a bit of a basic question!

Thanks again,
Matthew
________________________________________
From: Nikhil Kaza [nikhil.list@gmail.com]
Sent: Wednesday, December 16, 2009 4:14 PM
To: Petr PIKAL
Cc: whitaker m. (mw1006); r-help@r-project.org
Subject: Re: [R] Odp: Creating Dummy Variables in R

I don't think R will complain, if you use the approach below. However,
IF, VVS1 and VVS2 are linearly dependent.
Better use the factor approach and define which factor should be the
contrast

Nikhil

On 16 Dec 2009, at 10:12AM, Petr PIKAL wrote:

> what commands did you use for regression
>
> I suppose
>
> lm(Price~Weight+IF+VVS1+VVS2, data=your.data)
>
> shall not complain if your.data is a data frame.
>
> Regards
> Petr

== 2 of 4 ==
Date: Wed, Dec 16 2009 2:44 pm
From: "Nordlund, Dan (DSHS/RDA)"

> -----Original Message-----
> From: r-help-bounces@r-project.org [mailto:r-help-bounces@r-project.org] On
> Behalf Of whitaker m. (mw1006)
> Sent: Wednesday, December 16, 2009 2:14 PM
> To: Nikhil Kaza; Petr PIKAL
> Cc: r-help@r-project.org
> Subject: Re: [R] Odp: Creating Dummy Variables in R
>
> I have a much larger dataset than in my original email (attached - price dependent
> upon weight, Clarity (different levels IF-SI2), colour(levels D-L) and Cut (ideal-fair),
> and tried the regression command:
>
> >diamond.lm<-
> lm(price~weight+IF+VVS1+VVS2+VS1+VS2+SI1+SI2+I1+I2+D+E+F+G+H+I+J+K
> +L+ideal+excellent+very.good+good+fair, data="Diamonds2.txt")
>
> Error in eval(predvars, data, env) : invalid 'envir' argument
>
> Which lead to the error message below the command. I have tried searching for
> this, and assumed this was down to having categrocial variables within the data, is
> this a correct assumption or am i doing something else wrong? Apologies if this is a
> bit of a basic question!
>
> Thanks again,
> Matthew

You need to read your data from Diamonds2.txt into a dataframe first before running the lm() function. What does your file Diamonds2.txt look like?

Dan

Daniel J. Nordlund
Washington State Department of Social and Health Services
Planning, Performance, and Accountability
Research and Data Analysis Division
Olympia, WA 98504-5204

== 3 of 4 ==
Date: Wed, Dec 16 2009 2:57 pm
From: Peter Dalgaard

Nordlund, Dan (DSHS/RDA) wrote:
>> -----Original Message-----
>> From: r-help-bounces@r-project.org [mailto:r-help-bounces@r-project.org] On
>> Behalf Of whitaker m. (mw1006)
>> Sent: Wednesday, December 16, 2009 2:14 PM
>> To: Nikhil Kaza; Petr PIKAL
>> Cc: r-help@r-project.org
>> Subject: Re: [R] Odp: Creating Dummy Variables in R
>>
>> I have a much larger dataset than in my original email (attached - price dependent
>> upon weight, Clarity (different levels IF-SI2), colour(levels D-L) and Cut (ideal-fair),
>> and tried the regression command:
>>
>>> diamond.lm<-
>> lm(price~weight+IF+VVS1+VVS2+VS1+VS2+SI1+SI2+I1+I2+D+E+F+G+H+I+J+K
>> +L+ideal+excellent+very.good+good+fair, data="Diamonds2.txt")
>>
>> Error in eval(predvars, data, env) : invalid 'envir' argument
>>
>> Which lead to the error message below the command. I have tried searching for
>> this, and assumed this was down to having categrocial variables within the data, is
>> this a correct assumption or am i doing something else wrong? Apologies if this is a
>> bit of a basic question!
>>
>> Thanks again,
>> Matthew
>
> You need to read your data from Diamonds2.txt into a dataframe first before running the lm() function. What does your file Diamonds2.txt look like?

And, to put it more bluntly, he needs to study some introductory R text
rather more carefully to learn how the pieces fit together.

(Although I can see that the error may be cryptic to a beginner, I am at
a loss to explain what kind of leap of logic led him to believe that
dummy variables had _anything_ to do with it. For Heaven's sake, he must
be getting the same error if he leaves out the dummy variables!)

--
O__ ---- Peter Dalgaard Øster Farimagsgade 5, Entr.B
c/ /'_ --- Dept. of Biostatistics PO Box 2099, 1014 Cph. K
(*) \(*) -- University of Copenhagen Denmark Ph: (+45) 35327918
~~~~~~~~~~ - (p.dalgaard@biostat.ku.dk) FAX: (+45) 35327907

== 4 of 4 ==
Date: Wed, Dec 16 2009 3:10 pm
From: Rolf Turner

On 17/12/2009, at 11:14 AM, whitaker m. (mw1006) wrote:

> I have a much larger dataset than in my original email (attached -
> price dependent upon weight, Clarity (different levels IF-SI2),
> colour(levels D-L) and Cut (ideal-fair), and tried the regression
> command:
>
>> diamond.lm<-lm(price~weight+IF+VVS1+VVS2+VS1+VS2+SI1+SI2+I1+I2+D+E
>> +F+G+H+I+J+K+L+ideal+excellent+very.good+good+fair,
>> data="Diamonds2.txt")
>
> Error in eval(predvars, data, env) : invalid 'envir' argument
>
> Which lead to the error message below the command. I have tried
> searching for this, and assumed this was down to having categrocial
> variables within the data, is this a correct assumption or am i
> doing something else wrong? Apologies if this is a bit of a basic
> question!

(a) You don't want the quote marks around the data argument. That is
the source
of the "invalid 'envir' argument" error.

(b) You are not using the power of R. ***Don't*** create your own
dummy variables;
let lm() do it for you. Learn something about how R works, for
crying out loud.

Essentially you should be doing something like

diamond.lm <- lm(price ~ weight + Clarity + colour + Cut, data =
Diamond.txt)

where price, weight, Clarity, colour, and Cut are columns of the data
frame
Diamond.txt. The columns price and weight should be numeric vectors;
Clarity,
colour, and Cut should be ***factors***.

It is slightly worrying that you refer to ``Diamond.txt''. That
``.txt'' suffix
would lead me to believe that ``Diamond.txt'' is a (text) file
containing your
data. If that is the case, this won't work. The ``data'' argument
to lm() must
be an ***R object***. You have to read the data file into an R
object before trying
to use the data in a call to lm(). Something like

Diamond <- read.table("Diamond.txt") # Note that you ***do*** want
to quote the file name.

Then

diamond.lm <- lm(price ~ weight + Clarity + colour + Cut, data =
Diamond)

should do what you want. The dummy variable encoding used will be
determined
by the (first) value of options()$contrasts, which by default i
contr.treatment.

Read up on factors and contrasts.

cheers,

Rolf Turner

######################################################################
Attention:\ This e-mail message is privileged and confid...{{dropped:9}}

==============================================================================
TOPIC: Stata files
http://groups.google.com/group/r-help-archive/t/db776a5252384a5b?hl=en
==============================================================================

== 1 of 1 ==
Date: Wed, Dec 16 2009 3:12 pm
From: David Scott

Thank you both (JRG and BDR) for the advice. Just to clarify Brian, I
don't actually have any example .dta files yet to test. I wanted to be
prepared and be able to advise the client what to do.

I don't see anything about 'Stata 11 help for dta' Brian: must have a
different URL. But I did note already that Stata 11 uses format 114
which I agree indicates 10 and 11 use the same format.

David

JRG wrote:
> On 16 Dec 2009 at 10:24, Prof Brian Ripley wrote:
>
>> Have you tried read.dta?
>>
>> The URL on the help page for read.dta now says 'Stata 11 help for
>> dta', but it seems to document the same format (114) as Stata 10 (see
>> the value section). So in so far as Stata is telling us, it looks
>> like Stata 10 and 11 use the same format (as Stata 8 and 9 did).
>>
>
> That's correct: Stata 10 and Stata 11 use the same file format. If you need a still older format,
> the -saveold- command in Version 11 of Stata will save a Version 10/11 file in the format that was
> used by Versions 8 and 9 of Stata.
>
> ---JRG
>
>
>> On Wed, 16 Dec 2009, David Scott wrote:
>>
>>> I have a client who uses Stata 11.
>>>
>>> Can anyone advise me on ways of transferring data from this version of Stata
>>> to R?
>>>
>>> Reading the documentation for read.dta in package foreign it seems it only
>>> allows for versions up to Stata 10. I don't know Stata at all but a bit of
>>> poking around on their website led me to believe Stata 11 couldn't write in
>>> Stata 10 format. However Stata 11 can write Excel files in XML, so presumably
>>> .xlsx files.
>> I suspect it can write .csv files. See for example
>> http://www.stata.com/help.cgi?outsheet
>>
>>
>>> Any advice, correction of my misconceptions, etc. gladly received.
>>>
>>> David Scott
>>>
>>> _________________________________________________________________
>>> David Scott Department of Statistics
>>> The University of Auckland, PB 92019
>>> Auckland 1142, NEW ZEALAND
>>> Phone: +64 9 923 5055, or +64 9 373 7599 ext 85055
>>> Email: d.scott@auckland.ac.nz, Fax: +64 9 373 7018
>>>
>>> Director of Consulting, Department of Statistics
>> --
>> Brian D. Ripley, ripley@stats.ox.ac.uk
>> Professor of Applied Statistics, http://www.stats.ox.ac.uk/~ripley/
>> University of Oxford, Tel: +44 1865 272861 (self)
>> 1 South Parks Road, +44 1865 272866 (PA)
>> Oxford OX1 3TG, UK Fax: +44 1865 272595
>>
>> ______________________________________________
>> R-help@r-project.org mailing list
>> https://stat.ethz.ch/mailman/listinfo/r-help
>> PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
>> and provide commented, minimal, self-contained, reproducible code.
>
>
> John R. Gleason
> Associate Professor
>
> Syracuse University
> 430 Huntington Hall Voice: 315-443-3107
> Syracuse, NY 13244-2340 USA FAX: 315-443-4085
>
> PGP public key at keyservers
>
> ______________________________________________
> R-help@r-project.org mailing list
> https://stat.ethz.ch/mailman/listinfo/r-help
> PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
> and provide commented, minimal, self-contained, reproducible code.

--
_________________________________________________________________
David Scott Department of Statistics
The University of Auckland, PB 92019
Auckland 1142, NEW ZEALAND
Phone: +64 9 923 5055, or +64 9 373 7599 ext 85055
Email: d.scott@auckland.ac.nz, Fax: +64 9 373 7018

Director of Consulting, Department of Statistics

== 1 of 1 ==
Date: Wed, Dec 16 2009 7:16 pm
From: "rusers.sh"

Hi,
I have installed the necessary tools for making a R package under windows
and am sure these tool have been correctly configured. I am very new to try
writing a package, so the error may be very obvious for you. Forgive me if
it is too easy.
My package name is *stam*, and i put it under the folder
"D:/StatSoft/R/MyPackage/". I have two folders,*R* and *man*. Besides, i
also have *description* and *namespace** *files. It is a very simple beta
package.
Following is the error after using "R CMD check mypackage". Thanks very
much.
#########
* using log directory 'D:/StatSoft/R/MyPackage/stam.Rcheck'
* using R version 2.10.0 (2009-10-26)
* using session charset: ISO8859-1
* checking for file 'stam/DESCRIPTION' ... OK
* this is package 'stam' version '1.0-0'
* checking package name space information ... OK
* checking package dependencies ... OK
* checking if this is a source package ... OK
* checking for executable files ... OK
* checking whether package 'stam' can be installed ... OK
* checking package directory ... OK
* checking for portable file names ... OK
* checking DESCRIPTION meta-information ... OK
* checking top-level files ... OK
* checking index information ... OK
* checking package subdirectories ... OK
* checking R files for non-ASCII characters ... OK
* checking R files for syntax errors ... OK
* checking whether the package can be loaded ... ERROR
Error in library.dynam(lib, package, package.lib) :
shared library 'stam' not found
Error: package/namespace load failed for 'stam'
Execution halted

It looks like this package has a loading problem: see the messages for
details.

#########

--
-----------------
Jane Chang
Queen's

[[alternative HTML version deleted]]

== 1 of 1 ==
Date: Wed, Dec 16 2009 7:22 pm
From: Roslina Zakaria

Hi,

I would like to change a scale of the grid lines in a plot. I can see the default is 0.2, so how do I change to 0.1?

Thank you so much for any help given.

[[alternative HTML version deleted]]

== 1 of 2 ==
Date: Wed, Dec 16 2009 7:55 pm
From: gcam

Hi all.

So I have a data frame with multiple columns/variables. The first variable
is a major sample name for which there are some sub-samples. Currently I
have used the following command to remove the duplicates:

Samps_working<-Samps[-c(which(duplicated(Samps$ESR_Ref_edit))),]

This removes all of the duplicated sample rows.

However, I just realised that, of course, this removes the first observation
of each duplicated set. However, I wish to retain any that have the code
"Y" in another variable Samps$Loaded. I'm at a bit of a loss as to how best
to approach this problem.

Just to reiterate. I want to remove all duplicate lines based on sample
name, but, I want the lines to be removed with a preference given to those
that do not include a "Y" in the Loaded variable column.
--
View this message in context: http://n4.nabble.com/Remove-duplicates-from-a-data-frame-but-with-some-special-requirements-tp965745p965745.html
Sent from the R help mailing list archive at Nabble.com.

== 2 of 2 ==
Date: Wed, Dec 16 2009 9:08 pm
From: Gray Calhoun

Hi,
Try:

subset(Samps, !duplicated(Samps$ESR_ref_edit) | Samps$Loaded == "Y")

I'd need specific code to be sure that this is exactly what you want
(ie you specify input and desired output), but indexing with a logical
vector is probably going to be the solution.

Best,
Gray

--
Gray Calhoun

Assistant Professor of Economics
Iowa State University

==============================================================================

You received this message because you are subscribed to the Google Groups "R-help-archive"
group.

To post to this group, send email to r-help@r-project.org or visit http://groups.google.com/group/r-help-archive?hl=en

To unsubscribe from this group, send email to r-help-archive+unsubscribe@googlegroups.com

To change the way you get mail from this group, visit:
http://groups.google.com/group/r-help-archive/subscribe?hl=en

To report abuse, send email explaining the problem to abuse@googlegroups.com

==============================================================================
Google Groups: http://groups.google.com/?hl=en

r-help-archive - 25 new messages in 13 topics - digest

2009-12-16T11:14:00.000-08:00

R-help-archive
http://groups.google.com/group/r-help-archive?hl=en

r-help-archive@googlegroups.com

Today's topics:

== 1 of 1 ==
Date: Wed, Dec 16 2009 5:39 am
From: Terry Therneau

--- begin included message ---
Hi,

I am trying to write my own split function for rpart. The aim is to do,
instead of anova, a linear regression to determine the split (minimize
some criterion like sum of rss left and right of the split). The
regression (lm) should simply use the dependent and independent
variables passed to rpart.

---- end inclusion -----------

The rpart routine provides the x variables to a user-written split
function one at a time. Since the entire structure of rpart --
printing, plotting, tree representation, etc --- is based on the premise
of a single variable driving each split, what you are asking for would
require an entirely different program.

Terry Therneau

==============================================================================
TOPIC: rcart - classification and regression trees (CART)
http://groups.google.com/group/r-help-archive/t/e14c6769fcfa826a?hl=en
==============================================================================

== 1 of 2 ==
Date: Wed, Dec 16 2009 5:44 am
From: Frank E Harrell Jr

Katie N wrote:
> Hi,
> I am trying to use CART to find an ideal cut-off value for a simple
> diagnostic test (ie when the test score is above x, diagnose the condition).
> When I put in the model
>
> fit=rpart(outcome ~ predictor1(TB144), method="class", data=data8)
>
> sometimes it gives me a tree with multiple nodes for the same predictor (see
> below for example of tree with 1 or multiple nodes). Is there a way to tell
> it to make only 1 node? Or is it safe to assume that the cut-off value on
> the primary node is the ideal cut-off?
>
> Thanks!
> Katie
>
> http://n4.nabble.com/file/n964970/smartDNA%2BCART%2B-%2BTB144n.jpg
>
> http://n4.nabble.com/file/n964970/smartDNA%2BCART%2B-%2BTB122n.jpg
>
>

Katie,

Do note that the strategy you are using is inconsistent with decision
theory. Optimal decisions have to condition on everything you know
about a single patient, and do not ask the question "to what group does
this patient belong?". For example, we estimate something given the
patient's age is 20 instead of given that her age is less than 60.
That's why logistic regression is used so frequently to estimate
probabilities of disease. Any cutoff that must be used has to be on the
predicted probability scale in order to get an optimum decision, and
that cutoff must be specified by the provider of the utility function.
Even then the cutoff is not fully trusted, e.g., a physician may order
another test as the last minute when the probability of disease is in a
gray zone.

Frank
--
Frank E Harrell Jr Professor and Chair School of Medicine
Department of Biostatistics Vanderbilt University

== 2 of 2 ==
Date: Wed, Dec 16 2009 6:06 am
From: Terry Therneau

"Is there a way to tell it to make a tree with only one node?"
- see the maxdepth parameter in ?rpart.control

"Is it safe to assume that the cut-off value on the primary node is the
ideal cut-off?"

- trees are built sequentially; the first split will be the same for a
tree with only one split or one that continues further.

Terry T.

==============================================================================
TOPIC: Changing Column names in (Output) csv file
http://groups.google.com/group/r-help-archive/t/2eabb25c696f8d89?hl=en
==============================================================================

== 1 of 1 ==
Date: Wed, Dec 16 2009 5:46 am
From: Nikhil Kaza

I am not sure why you are reading and writing from disk so many times?
It will degrade performance.
Also avoid loops when you can.

just use

ONS$labels <- paste(ONS$Var1[i], ONS$Var2[i], ONS$Var3[i], ONS
$Var4[i], ONS$Var5[i], ONS$Var6[i], ONS$Var7[i], ONS$Var8[i],ONS
$Var9[i], ONS$Var10[i])
write.csv(c(ONS$labels, ONS$Freq), file="prob_table.csv")

Alternately just use concatenate command in excel after you exported
the joint probs.

Nikhil

On 15 Dec 2009, at 7:53AM, Amelia Livington wrote:

>
> Dear R helpers
>
> Following is a part of R code.
>
> data_lab <- expand.grid(c("R11", "R12", "R13"), c("R21", "R22",
> "R23"), c("R31", "R32", "R33"), c("R41", "R42", "R43"), c("R51",
> "R52", "R53"), c("R61", "R62", "R63"), c("R71", "R72", "R73"),
> c("R81", "R82", "R83"), c("R91", "R92", "R93"), c("R101", "R102",
> "R103"))
>
> range_prob <- list()
> range_prob[[1]] <- c(0.42,0.22,0.36)
> range_prob[[2]] <- c(0.14,0.56,0.30)
> range_prob[[3]] <- c(0.61,0.38,0.01)
> range_prob[[4]] <- c(0.34,0.37,0.29)
> range_prob[[5]] <- c(0.09,0.19,0.72)
> range_prob[[6]] <- c(0.42,0.21,0.37)
> range_prob[[7]] <- c(0.44,0.07,0.49)
> range_prob[[8]] <- c(0.54,0.06,0.40)
> range_prob[[9]] <- c(0.26,0.62,0.12)
> range_prob[[10]] <- c(0.65,0.19,0.16)
>
>
> pdf <- expand.grid(range_prob)
>
> data_lab$probs <- apply(pdf, 1, prod)
>
> joint_probs = xtabs(probs ~ Var1 +
> Var2+Var3+Var4+Var5+Var6+Var7+Var8+Var9+Var10, data = data_lab)
>
> write.csv(data.frame(joint_probs), 'joint_probs.csv', row.names =
> FALSE)
>
> ONS = read.csv('joint_probs.csv')
>
> Names = NULL
>
> for (i in 1:length(joint_probs))
> {
> Names[i] = paste(ONS$Var1[i], ONS$Var2[i], ONS$Var3[i], ONS
> $Var4[i], ONS$Var5[i], ONS$Var6[i], ONS$Var7[i], ONS$Var8[i],ONS
> $Var9[i], ONS$Var10[i])
> }
>
> write.csv(data.frame(labels = Names), 'Names.csv', row.names = FALSE)
>
> result = data.frame(read.csv('Names.csv')$labels,
> read.csv('joint_probs.csv')$Freq)
>
> write.csv(data.frame(result), 'prob_table.csv', row.names = FALSE)
>
> # The PROBLEM
>
> When I open the prob_table.csv file in Excel, instead of having
> column names as lables and Freq, I get the column heads as
>
> read.csv..Names.csv...labels
> read.csv..joint_probs.csv...Freq
> R11 R21 R31 R41 R51 R61 R71 R81 R91 R101 1.85E - 5
>
> and so on.
>
> Ideally I will like to have the column names as
>
> Label Probability
>
>
>
> Please guide
>
> Amelia
>
>
>
>
> [[alternative HTML version deleted]]
>
> ______________________________________________
> R-help@r-project.org mailing list
> https://stat.ethz.ch/mailman/listinfo/r-help
> PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
> and provide commented, minimal, self-contained, reproducible code.

==============================================================================
TOPIC: How to find the significant digits of a number?
http://groups.google.com/group/r-help-archive/t/523f040bf1c40f19?hl=en
==============================================================================

== 1 of 1 ==
Date: Wed, Dec 16 2009 5:56 am
From: Duncan Murdoch

On 16/12/2009 8:48 AM, Xiang Wu wrote:
> Yes, that's true. But what I need is the original precision of a
> numeric. So to me, 1.00 and 1.0000 are different.

In that case, they are not numbers, but strings: and you can probably
use nchar() to count characters, after stripping off any leading or
trailing whitespace, commas, and decimal points.

Duncan Murdoch

>
>
> On Wed, Dec 16, 2009 at 6:31 AM, Duncan Murdoch <murdoch@stats.uwo.ca
> <mailto:murdoch@stats.uwo.ca>> wrote:
>
> On 16/12/2009 12:39 AM, Xiang Wu wrote:
>
> Is there a function in R that could find the significant digit
> of a specific
> number? Such as for 3.1415, return '5'?
>
>
> The question may not have an unambiguous answer: I would say 1.00
> and 1.00000 have a different number of significant digits, but once
> converted to numbers they are identical in R.
>
> Duncan Murdoch
>
>

==============================================================================
TOPIC: Structural Equation Models(SEM)
http://groups.google.com/group/r-help-archive/t/1972caa3c14007b1?hl=en
==============================================================================

== 1 of 1 ==
Date: Wed, Dec 16 2009 6:25 am
From: "John Fox"

Dear Jarrett et al.,

I apologize for missing Ralf Finne's original posting -- I was out of town
at the time.

A few comments on this code:

(1) You should be able to get standard errors from the GLS solution (though
there's a missing factor 0.5 in the definition of objective.2 that should
deflate the standard errors by a factor of sqrt(2), and unfortunately this
is what one would get by default), but other output from summary.sem() won't
be meaningful.

(2) The difference between objective.1 and objective.2 in the original code
for sem.default() is that the latter uses an analytic gradient. The
implementation here of objective.2 for GLS and ULS fits doesn't supply a
gradient and consequently it would be clearer just to use objective.2 <-
objective.1 for these cases. Even better would be to set analytic.gradient
to FALSE in these cases. Of course, one could figure out the gradient for
the GLS and ULS estimators, but that's probably not worth the effort.

(3) Most of the output from summary(), etc., including standard errors, for
the ULS fit won't be sensible. The ULS "fitting function" never made sense
to me since it's not invariant with respect to rescaling of the observed
variables, but perhaps I'm missing something.

(4) I think that it would make most sense to save the fitting function in
the object returned by sem(), and then to tailor the output of other
functions such as summary.sem() and vcov.sem() to each case.

Regards,
John

> -----Original Message-----
> From: r-help-bounces@r-project.org [mailto:r-help-bounces@r-project.org]
On
> Behalf Of Jarrett Byrnes
> Sent: December-15-09 5:07 PM
> To: r-help@r-project.org
> Subject: Re: [R] Structural Equation Models(SEM)
>
> Joerg Everman has a great solution to this. He changed the middle of
> the sem.mod code to include a variable, fit, and then used the
> following approach around where you define the objectives:
>
> if (fit=="ml") {
> objective.1 <- function(par){
> A <- P <- matrix(0, m, m)
> val <- ifelse (fixed, ram[,5], par[sel.free])
> A[arrows.1] <- val[one.head]
> P[arrows.2t] <- P[arrows.2] <- val[!one.head]
> I.Ainv <- solve(diag(m) - A)
> C <- J %*% I.Ainv %*% P %*% t(I.Ainv) %*% t(J)
> Cinv <- solve(C)
> f <- sum(diag(S %*% Cinv)) + log(det(C))
> attributes(f) <- list(C=C, A=A, P=P)
> f
> }
> objective.2 <- function(par){
> A <- P <- matrix(0, m, m)
> val <- ifelse (fixed, ram[,5], par[sel.free])
> A[arrows.1] <- val[one.head]
> P[arrows.2t] <- P[arrows.2] <- val[!one.head]
> I.Ainv <- solve(diag(m) - A)
> C <- J %*% I.Ainv %*% P %*% t(I.Ainv) %*% t(J)
> Cinv <- solve(C)
> f <- sum(diag(S %*% Cinv)) + log(det(C))
> grad.P <- correct * t(I.Ainv) %*% t(J) %*% Cinv %*% (C - S) %*
> % Cinv %*% J %*% I.Ainv
> grad.A <- grad.P %*% P %*% t(I.Ainv)
> gradient <- rep(0, t)
> gradient[unique.free.1] <- tapply(grad.A[arrows.
> 1.free],ram[ram[,1]==1 & ram[,4]!=0, 4], sum)
> gradient[unique.free.2] <- tapply(grad.P[arrows.
> 2.free],ram[ram[,1]==2 & ram[,4]!=0, 4], sum)
> attributes(f) <- list(C=C, A=A, P=P, gradient=gradient)
> f
> }
> }
> if (fit=="gls") {
> objective.1 <- function(par){
> A <- P <- matrix(0, m, m)
> val <- ifelse (fixed, ram[,5], par[sel.free])
> A[arrows.1] <- val[one.head]
> P[arrows.2t] <- P[arrows.2] <- val[!one.head]
> I.Ainv <- solve(diag(m) - A)
> C <- J %*% I.Ainv %*% P %*% t(I.Ainv) %*% t(J)
> Sinv <- solve(S)
> f <- 0.5 * sum(diag( ((S - C) %*% Sinv) %*% ((S - C) %*%
> Sinv) ))
> attributes(f) <- list(C=C, A=A, P=P)
> f
> }
> objective.2 <- function(par){
> A <- P <- matrix(0, m, m)
> val <- ifelse (fixed, ram[,5], par[sel.free])
> A[arrows.1] <- val[one.head]
> P[arrows.2t] <- P[arrows.2] <- val[!one.head]
> I.Ainv <- solve(diag(m) - A)
> C <- J %*% I.Ainv %*% P %*% t(I.Ainv) %*% t(J)
> Sinv <- solve(S)
> f <- sum(diag( ((S - C) %*% Sinv) %*% ((S - C) %*% Sinv) ))
> # grad.P <- correct * t(I.Ainv) %*% t(J) %*% Cinv %*% (C - S) %*
> % Cinv %*% J %*% I.Ainv
> # grad.A <- grad.P %*% P %*% t(I.Ainv)
> # gradient <- rep(0, t)
> # gradient[unique.free.1] <- tapply(grad.A[arrows.
> 1.free],ram[ram[,1]==1 & ram[,4]!=0, 4], sum)
> # gradient[unique.free.2] <- tapply(grad.P[arrows.
> 2.free],ram[ram[,1]==2 & ram[,4]!=0, 4], sum)
> # attributes(f) <- list(C=C, A=A, P=P, gradient=gradient)
> attributes(f) <- list(C=C, A=A, P=P)
> f
> }
> }
> if (fit=="uls") {
> objective.1 <- function(par){
> A <- P <- matrix(0, m, m)
> val <- ifelse (fixed, ram[,5], par[sel.free])
> A[arrows.1] <- val[one.head]
> P[arrows.2t] <- P[arrows.2] <- val[!one.head]
> I.Ainv <- solve(diag(m) - A)
> C <- J %*% I.Ainv %*% P %*% t(I.Ainv) %*% t(J)
> Sinv <- solve(S)
> f <- 0.5 * sum(diag( (S - C) %*% (S - C) ))
> attributes(f) <- list(C=C, A=A, P=P)
> f
> }
> objective.2 <- function(par){
> A <- P <- matrix(0, m, m)
> val <- ifelse (fixed, ram[,5], par[sel.free])
> A[arrows.1] <- val[one.head]
> P[arrows.2t] <- P[arrows.2] <- val[!one.head]
> I.Ainv <- solve(diag(m) - A)
> C <- J %*% I.Ainv %*% P %*% t(I.Ainv) %*% t(J)
> Sinv <- solve(S)
> f <- 0.5 * sum(diag( (S - C) %*% (S - C) ))
> # grad.P <- correct * t(I.Ainv) %*% t(J) %*% Cinv %*% (C - S) %*
> % Cinv %*% J %*% I.Ainv
> # grad.A <- grad.P %*% P %*% t(I.Ainv)
> # gradient <- rep(0, t)
> # gradient[unique.free.1] <- tapply(grad.A[arrows.
> 1.free],ram[ram[,1]==1 & ram[,4]!=0, 4], sum)
> # gradient[unique.free.2] <- tapply(grad.P[arrows.
> 2.free],ram[ram[,1]==2 & ram[,4]!=0, 4], sum)
> # attributes(f) <- list(C=C, A=A, P=P, gradient=gradient)
> attributes(f) <- list(C=C, A=A, P=P)
> f
> }
> }
>
>
> On Dec 2, 2009, at 10:22 AM, Jeremy Miles wrote:
>
> > In the world of SEM, GLS has pretty much fallen by the wayside - I
> > can't recall anything I've seen arguing for it's use in the past 10
> > years, and I also can't recall anyone using it over ML. The
> > recommendations for non-normal distributions tend to be robust-ML, or
> > robust weighted least squares. These are more computationally
> > intensive, and I *think* that John Fox (author of sem) has written
> > somewhere that it wouldn't be possible to implement them within R,
> > without using a lower level language - or rather that it might be
> > possible, but it would be really, really slow.
> >
> > However, ML and GLS are pretty similar, if you dug around in the
> > source code, you could probably make the change (see,
> > http://www2.gsu.edu/~mkteer/discrep.html for example, for the
> > equations; in fact GLS is somewhat computationally simpler, as you
> > don't need to invert the implied covariance matrix at each iteration).
> > However, the fact that it's not hard to make the change, and that no
> > one has made the change, is another argument that it's not a change
> > that needs to be made.
> >
> > Jeremy
> >
> >
> >
> > 2009/12/2 Ralf Finne <Ralf.Finne@novia.fi>:
> >> Hi R-colleagues.
> >>
> >> I have been using the sem(sem) function. It uses
> >> maximum likelyhood as optimizing. method.
> >> According to simulation study in Umeå Sweden
> >>
(http://www.stat.umu.se/kursweb/vt07/stad04mom3/?download=UlfHolmberg.pdf
> >> Sorry it is in swedish, except the abstract)
> >> maximum likelihood is OK for large samples and normal distribution
> >> the SEM-problem should be optimized by GLS (Generalized Least
> >> Squares).
> >>
> >>
> >> So to the question:
> >>
> >> Is there any R-function that solves SEM with GLS?
> >>
> >>
> >> Ralf Finne
> >> Novia University of Applied Science
> >> Vasa Finland
> >>
> >> ______________________________________________
> >> R-help@r-project.org mailing list
> >> https://stat.ethz.ch/mailman/listinfo/r-help
> >> PLEASE do read the posting guide http://www.R-project.org/posting-
> guide.html
> >> and provide commented, minimal, self-contained, reproducible code.
> >>
> >
> >
> >
> > --
> > Jeremy Miles
> > Psychology Research Methods Wiki: www.researchmethodsinpsychology.com
> >
> > ______________________________________________
> > R-help@r-project.org mailing list
> > https://stat.ethz.ch/mailman/listinfo/r-help
> > PLEASE do read the posting guide http://www.R-project.org/posting-
> guide.html
> > and provide commented, minimal, self-contained, reproducible code.
>
> ______________________________________________
> R-help@r-project.org mailing list
> https://stat.ethz.ch/mailman/listinfo/r-help
> PLEASE do read the posting guide
http://www.R-project.org/posting-guide.html
> and provide commented, minimal, self-contained, reproducible code.

==============================================================================
TOPIC: R and Hierarchical Forecasting
http://groups.google.com/group/r-help-archive/t/fb31ebad900d3580?hl=en
==============================================================================

== 1 of 1 ==
Date: Wed, Dec 16 2009 6:44 am
From: Kjetil Halvorsen

I don't know of any prepackaged functions to do this, but I have done
this with some success,
using compositional methods: forecasting the total, and then
forecasting the compositions.,, with compositional
data analysis.
The package compositions or robCompositions
might be usefull. Or not.

Kjetil B Halvorsen

On Tue, Dec 15, 2009 at 10:13 PM, zubin <binabina@bellsouth.net> wrote:
> Hello, does anyone know of any R routines capable of whats called
> Hierarchical Forecasting, reconciling the different hierarchies.
>
> Example: A top down forecast where the corporate forecast is created and
> then all the regions within the corporate entity are also forecasted,
> with the constraint they sum to the corporate forecast.
>
> ______________________________________________
> R-help@r-project.org mailing list
> https://stat.ethz.ch/mailman/listinfo/r-help
> PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
> and provide commented, minimal, self-contained, reproducible code.
>

==============================================================================
TOPIC: Creating Dummy Variables in R
http://groups.google.com/group/r-help-archive/t/c9746670ecbbf081?hl=en
==============================================================================

== 1 of 6 ==
Date: Wed, Dec 16 2009 6:58 am
From: "whitaker m. (mw1006)"

Hi,
I am trying to create a set of dummy variables to use within a multiple linear regression and am unable to find the codes within the manuals.

For example i have:
Price Weight Clarity
IF VVS1 VVS2
500 8 1 0 0
1000 5.2 0 0 1
864 3 0 1 0
340 2.6 0 0 1
90 0.5 1 0 0
450 2.3 0 1 0

Where price is dependent upon weight (single value in each observation) and clarity (split into three levels, IF, VVS1, VVS2).
I am having trouble telling the program that clarity is a set of 3 dummy variables and keep getting error messages, what is the correct way?

Any helps is greatly appreciated.
Matthew

== 2 of 6 ==
Date: Wed, Dec 16 2009 7:12 am
From: Petr PIKAL

r-help-bounces@r-project.org napsal dne 16.12.2009 15:58:56:

> Hi,
> I am trying to create a set of dummy variables to use within a multiple
linear
> regression and am unable to find the codes within the manuals.
>
> For example i have:
> Price Weight Clarity
> IF VVS1 VVS2
> 500 8 1 0 0
> 1000 5.2 0 0 1
> 864 3 0 1 0
> 340 2.6 0 0 1
> 90 0.5 1 0 0
> 450 2.3 0 1 0
>
> Where price is dependent upon weight (single value in each observation)
and
> clarity (split into three levels, IF, VVS1, VVS2).
> I am having trouble telling the program that clarity is a set of 3 dummy

> variables and keep getting error messages, what is the correct way?

Well, try to bribe it. Or ask what please it to break its resistance.

Seriously. What is a structure of your data in R.

?str

what commands did you use for regression

I suppose

lm(Price~Weight+IF+VVS1+VVS2, data=your.data)

shall not complain if your.data is a data frame.

Regards
Petr

>
> Any helps is greatly appreciated.
> Matthew
>
> ______________________________________________
> R-help@r-project.org mailing list
> https://stat.ethz.ch/mailman/listinfo/r-help
> PLEASE do read the posting guide
http://www.R-project.org/posting-guide.html
> and provide commented, minimal, self-contained, reproducible code.

== 3 of 6 ==
Date: Wed, Dec 16 2009 7:16 am
From: S Devriese

On 12/16/2009 03:58 PM, whitaker m. (mw1006) wrote:
> Hi,
> I am trying to create a set of dummy variables to use within a multiple linear regression and am unable to find the codes within the manuals.
>
> For example i have:
> Price Weight Clarity
> IF VVS1 VVS2
> 500 8 1 0 0
> 1000 5.2 0 0 1
> 864 3 0 1 0
> 340 2.6 0 0 1
> 90 0.5 1 0 0
> 450 2.3 0 1 0
>
> Where price is dependent upon weight (single value in each observation) and clarity (split into three levels, IF, VVS1, VVS2).
> I am having trouble telling the program that clarity is a set of 3 dummy variables and keep getting error messages, what is the correct way?
>

Without an example of your code, it's a bit difficult. But it might be
easier to use one variable "clarity" with three possible values (IF,
VVS1, VVS2), defined as a factor.
lm(Price ~ Weight + Clarity) should then do the trick (unless you
explicitly want to use a different dummy coding than the default)

Stephan

== 4 of 6 ==
Date: Wed, Dec 16 2009 7:19 am
From: Achim Zeileis

On Wed, 16 Dec 2009, whitaker m. (mw1006) wrote:

> Hi,
> I am trying to create a set of dummy variables to use within a multiple linear regression and am unable to find the codes within the manuals.
>
> For example i have:
> Price Weight Clarity
> IF VVS1 VVS2
> 500 8 1 0 0
> 1000 5.2 0 0 1
> 864 3 0 1 0
> 340 2.6 0 0 1
> 90 0.5 1 0 0
> 450 2.3 0 1 0
>
> Where price is dependent upon weight (single value in each observation) and clarity (split into three levels, IF, VVS1, VVS2).
> I am having trouble telling the program that clarity is a set of 3 dummy variables and keep getting error messages, what is the correct way?

You should code the categorical variable "Clarity" as a "factor" so that R
knows that this is a categorical variable and can deal with it
appropriately in subsequent computations such as summary() or lm().

Thus, I would recommend to store your data as

dat <- data.frame(
Price = c(500, 1000, 864, 340, 90, 450),
Weight = c(8, 5.2, 3, 2.6, 0.5, 2.3),
Clarity = c("IF", "VVS1", "VVS2")[c(1, 3, 2, 3, 1, 2)])

which yields, e.g.,

R> summary(dat)
Price Weight Clarity
Min. : 90.0 Min. :0.500 IF :2
1st Qu.: 367.5 1st Qu.:2.375 VVS1:2
Median : 475.0 Median :2.800 VVS2:2
Mean : 540.7 Mean :3.600
3rd Qu.: 773.0 3rd Qu.:4.650
Max. :1000.0 Max. :8.000

and then you can also do

R> lm(Price ~ Weight + Clarity, data = dat)

Call:
lm(formula = Price ~ Weight + Clarity, data = dat)

Coefficients:
(Intercept) Weight ClarityVVS1 ClarityVVS2
-45.05 80.01 490.02 403.00

or if you wish to choose a different coding

R> lm(Price ~ 0 + Weight + Clarity, data = dat)

Call:
lm(formula = Price ~ 0 + Weight + Clarity, data = dat)

Coefficients:
Weight ClarityIF ClarityVVS1 ClarityVVS2
80.01 -45.05 444.97 357.95

Some further reading of introductory material on linear regression in R
would be useful. Also look at ?lm, ?factor, ?model.matrix, ?contrasts etc.

hth,
Z

> Any helps is greatly appreciated.
> Matthew
>
> ______________________________________________
> R-help@r-project.org mailing list
> https://stat.ethz.ch/mailman/listinfo/r-help
> PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
> and provide commented, minimal, self-contained, reproducible code.
>
>

== 5 of 6 ==
Date: Wed, Dec 16 2009 8:14 am
From: Nikhil Kaza

I don't think R will complain, if you use the approach below. However,
IF, VVS1 and VVS2 are linearly dependent.
Better use the factor approach and define which factor should be the
contrast

Nikhil

On 16 Dec 2009, at 10:12AM, Petr PIKAL wrote:

> what commands did you use for regression
>
> I suppose
>
> lm(Price~Weight+IF+VVS1+VVS2, data=your.data)
>
> shall not complain if your.data is a data frame.
>
> Regards
> Petr

== 6 of 6 ==
Date: Wed, Dec 16 2009 8:21 am
From: "Tom Fletcher"

Is your variable Clarity a categorical with 4 levels? Thus, the need for
k-1 (3) dummies? Your error may be the result of creating k instead of
k-1 dummies, but can't be sure from the example.

In R, you don't have to (unless you really want to) explicitly create
separate variables. You can use the internal contrast functions.

See

?contr.treatment

Which is dummy coding by default. You can specify which group is the
reference group.

Alternatively, if you prefer effects coding, you can see
?contr.sum

There are others as well.

Tom Fletcher

-----Original Message-----
From: r-help-bounces@r-project.org [mailto:r-help-bounces@r-project.org]
On Behalf Of whitaker m. (mw1006)
Sent: Wednesday, December 16, 2009 8:59 AM
To: r-help@r-project.org
Subject: [R] Creating Dummy Variables in R

Hi,
I am trying to create a set of dummy variables to use within a multiple
linear regression and am unable to find the codes within the manuals.

For example i have:
Price Weight Clarity
IF VVS1 VVS2
500 8 1 0 0
1000 5.2 0 0 1
864 3 0 1 0
340 2.6 0 0 1
90 0.5 1 0 0
450 2.3 0 1 0

Where price is dependent upon weight (single value in each observation)
and clarity (split into three levels, IF, VVS1, VVS2).
I am having trouble telling the program that clarity is a set of 3 dummy
variables and keep getting error messages, what is the correct way?

Any helps is greatly appreciated.
Matthew

______________________________________________
R-help@r-project.org mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide
http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.

==============================================================================
TOPIC: difference between the meaning of MARGIN in sweep() and apply()
http://groups.google.com/group/r-help-archive/t/3ad3a142e475c183?hl=en
==============================================================================

== 1 of 2 ==
Date: Wed, Dec 16 2009 7:09 am
From: Levi Waldron

For example, subtracting 1:2 from the rows of a two-column matrix:

> t(apply(matrix(1:6,ncol=2),MARGIN=1,function(y) y - 1:2))
[,1] [,2]
[1,] 0 2
[2,] 1 3
[3,] 2 4
> sweep(matrix(1:6,ncol=2),MARGIN=2,1:2,FUN="-")
[,1] [,2]
[1,] 0 2
[2,] 1 3
[3,] 2 4

Is there a logic to this difference, or is it just a quirk of the history of
these functions?

I found one discussion on this topic, but without what I thought was a very
conclusive end:

http://finzi.psych.upenn.edu/Rhelp08/2009-March/191930.html

--
Levi Waldron
post-doctoral fellow
Jurisica Lab, Ontario Cancer Institute
Division of Signaling Biology
TMDT 9-304D
101 College Street
Toronto, Ontario M5G 1L7
(416)581-7453

[[alternative HTML version deleted]]

== 2 of 2 ==
Date: Wed, Dec 16 2009 10:04 am
From: "Charles C. Berry"

On Wed, 16 Dec 2009, Levi Waldron wrote:

> For example, subtracting 1:2 from the rows of a two-column matrix:
>
>> t(apply(matrix(1:6,ncol=2),MARGIN=1,function(y) y - 1:2))
> [,1] [,2]
> [1,] 0 2
> [2,] 1 3
> [3,] 2 4
>> sweep(matrix(1:6,ncol=2),MARGIN=2,1:2,FUN="-")
> [,1] [,2]
> [1,] 0 2
> [2,] 1 3
> [3,] 2 4
>
> Is there a logic to this difference, or is it just a quirk of the history of
> these functions?

Well

sweep(mat,MARGIN, apply(mat,MARGIN, applyfun), sweepfun)

does it for me.

> mat <- matrix(1:6,ncol=2)
> sweep( mat, 2, apply(mat, 2, mean),'-')
[,1] [,2]
[1,] -1 -1
[2,] 0 0
[3,] 1 1
>

Chuck

>
> I found one discussion on this topic, but without what I thought was a very
> conclusive end:
>
> http://finzi.psych.upenn.edu/Rhelp08/2009-March/191930.html
>
>
> --
> Levi Waldron
> post-doctoral fellow
> Jurisica Lab, Ontario Cancer Institute
> Division of Signaling Biology
> TMDT 9-304D
> 101 College Street
> Toronto, Ontario M5G 1L7
> (416)581-7453
>
> [[alternative HTML version deleted]]
>
> ______________________________________________
> R-help@r-project.org mailing list
> https://stat.ethz.ch/mailman/listinfo/r-help
> PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
> and provide commented, minimal, self-contained, reproducible code.
>

Charles C. Berry (858) 534-2098
Dept of Family/Preventive Medicine
E mailto:cberry@tajo.ucsd.edu UC San Diego
http://famprevmed.ucsd.edu/faculty/cberry/ La Jolla, San Diego 92093-0901

==============================================================================
TOPIC: Duration model with sample selection (or selectivity)
http://groups.google.com/group/r-help-archive/t/ad2c5f1c6a4c6b00?hl=en
==============================================================================

== 1 of 1 ==
Date: Wed, Dec 16 2009 8:19 am
From: "Changyou Sun"

Hello All,

I am interested in estimating a duration model (also known as survival
analysis or event-history analysis). I use an economic dataset. In
economics terms, the model is "duration model with sample selection (or
selectivity)." The time spell variable is only observed for a sample
that meets certain requirements so the sample is nonrandom. Does anybody
know any R package that can take care of this?

Thank you.

Edwin

==============================================================================
TOPIC: Rgraphviz installation
http://groups.google.com/group/r-help-archive/t/167c59e5f3fc7f01?hl=en
==============================================================================

== 1 of 2 ==
Date: Wed, Dec 16 2009 9:06 am
From: Paul Evans

Hi,

I wanted to install Rgraphviz. From the R GUI, I downloaded the package but when I tried to load it I got the following error message:
"This application has failed to start because libcdt-4.dll was not
found. Re-installing the application may fix this problem."

The R commands and errors are:
> utils:::menuInstallPkgs()
trying URL 'http://www.bioconductor.org/packages/2.5/bioc/bin/windows/contrib/2.10/Rgraphviz_1.24.0.zip'
Content type 'application/zip' length 969595 bytes (946 Kb)
opened URL
downloaded 946 Kb

package 'Rgraphviz' successfully unpacked and MD5 sums checked

The downloaded packages are in
C:\Users\pe\AppData\Local\Temp\RtmpoebdlH\downloaded_packages
> library(Rgraphviz)
Error in inDL(x, as.logical(local), as.logical(now), ...) :
unable to load shared library 'C:/Users/pe/Documents/R/win-library/2.10/Rgraphviz/libs/Rgraphviz.dll':
LoadLibrary failure: The specified module could not be found.

Error: package/namespace load failed for 'Rgraphviz'
My sessioninfo is:
> sessionInfo()
R version 2.10.0 (2009-10-26)
i386-pc-mingw32

locale:
[1] LC_COLLATE=English_United States.1252 LC_CTYPE=English_United States.1252 LC_MONETARY=English_United States.1252
[4] LC_NUMERIC=C LC_TIME=English_United States.1252

attached base packages:
[1] grid stats graphics grDevices utils datasets methods base

other attached packages:
[1] GOstats_2.12.0 Category_2.12.0 multtest_2.2.0 R.utils_1.2.4 R.oo_1.6.5 R.methodsS3_1.0.3 genefilter_1.28.0
[8] hgu133a.db_2.3.5 org.Hs.eg.db_2.3.6 GEOquery_2.8.0 RCurl_1.2-1 bitops_1.0-4.1 biomaRt_2.2.0 affy_1.24.1
[15] int.intact.db_1.1.1 int.geneint.db_1.1.1 PAnnBuilder_1.8.0 RpsiXML_1.6.0 hypergraph_1.18.0 XML_2.6-0 RBGL_1.20.0
[22] annotate_1.24.0 GO.db_2.3.5 RSQLite_0.7-3 DBI_0.2-4 AnnotationDbi_1.8.0 Biobase_2.6.0 graph_1.24.1

loaded via a namespace (and not attached):
[1] affyio_1.14.0 GSEABase_1.8.0 MASS_7.3-3 org.Sc.sgd.db_2.3.5 preprocessCore_1.8.0 splines_2.10.0 survival_2.35-7
[8] tools_2.10.0 xtable_1.5-6
>

What do I need to do to install and load this successfully?

thanks!

[[alternative HTML version deleted]]

== 2 of 2 ==
Date: Wed, Dec 16 2009 9:14 am
From: Martin Morgan

Hi Paul

Rgraphviz is a bioconductor package, so ask on the Bioconductor mailing
list.

http://bioconductor.org/docs/mailList.html

Install Bioconductor packages following the instructions here

http://bioconductor.org/docs/install/

Rgraphviz on Windows installed from binary (as you're doing below)
requires that your computer has the same version of graphviz as the
computer used to build Rgraphviz. The details are in the package README
file (inconveniently linked from here

http://bioconductor.org/packages/2.5/bioc/html/Rgraphviz.html

http://bioconductor.org/packages/2.5/bioc/src/contrib/Rgraphviz_1.24.0.tar.gz

Cutting and pasting from this file, the instructions are:

o. install Graphviz ***2.20.3.1*** Windows ***binary*** distribution from

http://www.graphviz.org/pub/graphviz/stable/windows/graphviz-2.20.3.1.msi

in a directory that we'll call $DIR. We use the following as an example:

C:\Graphviz2.20

If you install it in other dir, just use your dir correspondingly.

o. Add Graphviz' bin diretory to the user PATH variable, e.g.,

c:\Graphviz2.20\bin

o. Check the configuration by starting a new DOS window and displaying
the PATH environment variable, e.g.,

c:\> echo %PATH%

o. Use biocLite('Rgraphviz') to install Rgraphviz

Paul Evans wrote:
> Hi,
>
> I wanted to install Rgraphviz. From the R GUI, I downloaded the package but when I tried to load it I got the following error message:
> "This application has failed to start because libcdt-4.dll was not
> found. Re-installing the application may fix this problem."
>
> The R commands and errors are:
>
>> utils:::menuInstallPkgs()
>>
> trying URL 'http://www.bioconductor.org/packages/2.5/bioc/bin/windows/contrib/2.10/Rgraphviz_1.24.0.zip'
> Content type 'application/zip' length 969595 bytes (946 Kb)
> opened URL
> downloaded 946 Kb
>
> package 'Rgraphviz' successfully unpacked and MD5 sums checked
>
> The downloaded packages are in
> C:\Users\pe\AppData\Local\Temp\RtmpoebdlH\downloaded_packages
>
>> library(Rgraphviz)
>>
> Error in inDL(x, as.logical(local), as.logical(now), ...) :
> unable to load shared library 'C:/Users/pe/Documents/R/win-library/2.10/Rgraphviz/libs/Rgraphviz.dll':
> LoadLibrary failure: The specified module could not be found.
>
>
> Error: package/namespace load failed for 'Rgraphviz'
> My sessioninfo is:
>
>> sessionInfo()
>>
> R version 2.10.0 (2009-10-26)
> i386-pc-mingw32
>
> locale:
> [1] LC_COLLATE=English_United States.1252 LC_CTYPE=English_United States.1252 LC_MONETARY=English_United States.1252
> [4] LC_NUMERIC=C LC_TIME=English_United States.1252
>
> attached base packages:
> [1] grid stats graphics grDevices utils datasets methods base
>
> other attached packages:
> [1] GOstats_2.12.0 Category_2.12.0 multtest_2.2.0 R.utils_1.2.4 R.oo_1.6.5 R.methodsS3_1.0.3 genefilter_1.28.0
> [8] hgu133a.db_2.3.5 org.Hs.eg.db_2.3.6 GEOquery_2.8.0 RCurl_1.2-1 bitops_1.0-4.1 biomaRt_2.2.0 affy_1.24.1
> [15] int.intact.db_1.1.1 int.geneint.db_1.1.1 PAnnBuilder_1.8.0 RpsiXML_1.6.0 hypergraph_1.18.0 XML_2.6-0 RBGL_1.20.0
> [22] annotate_1.24.0 GO.db_2.3.5 RSQLite_0.7-3 DBI_0.2-4 AnnotationDbi_1.8.0 Biobase_2.6.0 graph_1.24.1
>
> loaded via a namespace (and not attached):
> [1] affyio_1.14.0 GSEABase_1.8.0 MASS_7.3-3 org.Sc.sgd.db_2.3.5 preprocessCore_1.8.0 splines_2.10.0 survival_2.35-7
> [8] tools_2.10.0 xtable_1.5-6
>
>
> What do I need to do to install and load this successfully?
>
> thanks!
>
>
>
> [[alternative HTML version deleted]]
>
> ______________________________________________
> R-help@r-project.org mailing list
> https://stat.ethz.ch/mailman/listinfo/r-help
> PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
> and provide commented, minimal, self-contained, reproducible code.
>

--
Martin Morgan
Computational Biology / Fred Hutchinson Cancer Research Center
1100 Fairview Ave. N.
PO Box 19024 Seattle, WA 98109

Location: Arnold Building M1 B861
Phone: (206) 667-2793

==============================================================================
TOPIC: Baseline survival estimate
http://groups.google.com/group/r-help-archive/t/19d2d8063246f50b?hl=en
==============================================================================

== 1 of 3 ==
Date: Wed, Dec 16 2009 9:15 am
From: Laura Bonnett

Dear R-help,

I am trying to obtain the baseline survival estimate of a fitted Cox model
(S_0 (t)). I know that previous posts have said use 'basehaz' but this
gives the baseline hazard function and not the baseline survival estimate.
Is there a way to obtain the baseline survival estimate or do I have to use
the formula which does something like S(t) = exp[- the integral from 0 to t
of h(u) du]?

Thank you for your assistance,

Laura

fit1 <-
coxph(Surv(tsecond/365,seccens)~stroke(smess1)+othnd(smess1)+relat(smess1)+asleep(smess1)+abeeg1(smess1)+treat(smess1),data=smess1)
basehaz(fit1)

where stroke is a function which creates a binary variable from the dataset
smess1 etc.

[[alternative HTML version deleted]]

== 2 of 3 ==
Date: Wed, Dec 16 2009 9:34 am
From: tlumley@u.washington.edu

survival = exp(-hazard)

-thomas

On Wed, 16 Dec 2009, Laura Bonnett wrote:

> Dear R-help,
>
> I am trying to obtain the baseline survival estimate of a fitted Cox model
> (S_0 (t)). I know that previous posts have said use 'basehaz' but this
> gives the baseline hazard function and not the baseline survival estimate.
> Is there a way to obtain the baseline survival estimate or do I have to use
> the formula which does something like S(t) = exp[- the integral from 0 to t
> of h(u) du]?
>
> Thank you for your assistance,
>
> Laura
>
> fit1 <-
> coxph(Surv(tsecond/365,seccens)~stroke(smess1)+othnd(smess1)+relat(smess1)+asleep(smess1)+abeeg1(smess1)+treat(smess1),data=smess1)
> basehaz(fit1)
>
> where stroke is a function which creates a binary variable from the dataset
> smess1 etc.
>
> [[alternative HTML version deleted]]
>
> ______________________________________________
> R-help@r-project.org mailing list
> https://stat.ethz.ch/mailman/listinfo/r-help
> PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
> and provide commented, minimal, self-contained, reproducible code.
>

Thomas Lumley Assoc. Professor, Biostatistics
tlumley@u.washington.edu University of Washington, Seattle

== 3 of 3 ==
Date: Wed, Dec 16 2009 10:01 am
From: Marc Schwartz

I might just extend Thomas' succinct reply with a pointer to ?
survfit.coxph in the 'survival' package, which will provide the
baseline survival estimates at the mean values of the covariates from
the Cox model. It also has nice summary and plot methods and allows
for estimating predicted survival for given covariate values.

Using the 'ovarian' dataset:

fit <- coxph(Surv(futime, fustat) ~ age, data = ovarian)

# Note the default 'centered = TRUE' for basehaz()
> exp(-basehaz(fit)$hazard)
[1] 0.9880206 0.9738738 0.9545899 0.9334790 0.8973620 0.8624781
[7] 0.8243117 0.7750981 0.7244924 0.6734146 0.5962187 0.5204807

> summary(survfit(fit))$surv
[1] 0.9880206 0.9738738 0.9545899 0.9334790 0.8973620 0.8624781
[7] 0.8243117 0.7750981 0.7244924 0.6734146 0.5962187 0.5204807

You can also use:

summary(survfit(fit))
plot(survfit(fit))
...

You might also be interested in John Fox' web appendix on Cox models:

http://cran.r-project.org/doc/contrib/Fox-Companion/appendix-cox-regression.pdf

HTH,

Marc Schwartz

On Dec 16, 2009, at 11:34 AM, tlumley@u.washington.edu wrote:

>
>
> survival = exp(-hazard)
>
> -thomas
>
> On Wed, 16 Dec 2009, Laura Bonnett wrote:
>
>> Dear R-help,
>>
>> I am trying to obtain the baseline survival estimate of a fitted
>> Cox model
>> (S_0 (t)). I know that previous posts have said use 'basehaz' but
>> this
>> gives the baseline hazard function and not the baseline survival
>> estimate.
>> Is there a way to obtain the baseline survival estimate or do I
>> have to use
>> the formula which does something like S(t) = exp[- the integral
>> from 0 to t
>> of h(u) du]?
>>
>> Thank you for your assistance,
>>
>> Laura
>>
>> fit1 <-
>> coxph(Surv(tsecond/
>> 365
>> ,seccens
>> )~
>> stroke
>> (smess1
>> )
>> +
>> othnd
>> (smess1
>> )
>> +
>> relat
>> (smess1)+asleep(smess1)+abeeg1(smess1)+treat(smess1),data=smess1)
>> basehaz(fit1)
>>
>> where stroke is a function which creates a binary variable from the
>> dataset
>> smess1 etc.

==============================================================================
TOPIC: problem with a densityplot
http://groups.google.com/group/r-help-archive/t/db0e998618ec816a?hl=en
==============================================================================

== 1 of 3 ==
Date: Wed, Dec 16 2009 9:19 am
From: "ch3o@autistici.org"

Hi,
i have a script how i launch lattice to make a densityplot.
in the script:

jpeg(file="XXX.jpg")
densityplot(~f_diametro+m_diametro+n_diametro, plot.points="rug",
auto.key=T)
dev.off()

does'nt work and in R i dont have any output
but if i launch by R row by row, runs correctly.....
any idea?

== 2 of 3 ==
Date: Wed, Dec 16 2009 10:46 am
From: baptiste auguie

FAQ:
http://cran.r-project.org/doc/FAQ/R-FAQ.html#Why-do-lattice_002ftrellis-graphics-not-work_003f

you need to print()

HTH,

baptiste

2009/12/16 ch3o@autistici.org <ch3o@autistici.org>:
> Hi,
> i have a script how i launch lattice to make a densityplot.
> in the script:
>
> jpeg(file="XXX.jpg")
> densityplot(~f_diametro+m_diametro+n_diametro, plot.points="rug",
> auto.key=T)
> dev.off()
>
> does'nt work and in R i dont have any output
> but if i launch by R row by row, runs correctly.....
> any idea?
>
> ______________________________________________
> R-help@r-project.org mailing list
> https://stat.ethz.ch/mailman/listinfo/r-help
> PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
> and provide commented, minimal, self-contained, reproducible code.
>

== 3 of 3 ==
Date: Wed, Dec 16 2009 10:50 am
From: Benilton Carvalho

instead of densityplot(...) use

print(densityplot(...))

On Dec 16, 2009, at 3:19 PM, ch3o@autistici.org wrote:

> Hi,
> i have a script how i launch lattice to make a densityplot.
> in the script:
>
> jpeg(file="XXX.jpg")
> densityplot(~f_diametro+m_diametro+n_diametro, plot.points="rug",
> auto.key=T)
> dev.off()
>
> does'nt work and in R i dont have any output
> but if i launch by R row by row, runs correctly.....
> any idea?
>
> ______________________________________________
> R-help@r-project.org mailing list
> https://stat.ethz.ch/mailman/listinfo/r-help
> PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
> and provide commented, minimal, self-contained, reproducible code.

== 1 of 1 ==
Date: Wed, Dec 16 2009 11:11 am
From: baptiste auguie

Dear list,

library(rfaq)
faq("lattice")
## or if I were a number kind of person
faq(722)

The database could be split into "windows only", etc.

Just a thought anyway,

Best regards,

baptiste

==============================================================================

You received this message because you are subscribed to the Google Groups "R-help-archive"
group.

To post to this group, send email to r-help@r-project.org or visit http://groups.google.com/group/r-help-archive?hl=en

To unsubscribe from this group, send email to r-help-archive+unsubscribe@googlegroups.com

To change the way you get mail from this group, visit:
http://groups.google.com/group/r-help-archive/subscribe?hl=en

To report abuse, send email explaining the problem to abuse@googlegroups.com

==============================================================================
Google Groups: http://groups.google.com/?hl=en

r-help-archive - 25 new messages in 11 topics - digest

2009-12-16T05:40:00.000-08:00

R-help-archive
http://groups.google.com/group/r-help-archive?hl=en

r-help-archive@googlegroups.com

Today's topics:

==============================================================================
TOPIC: lasso regression coefficients
http://groups.google.com/group/r-help-archive/t/934b7123088cc14a?hl=en
==============================================================================

== 1 of 1 ==
Date: Wed, Dec 16 2009 12:55 am
From: Eleni Christodoulou

Dear list,

I have been trying to apply a simple lasso regression on a 10-element
vector, just to see how this method works so as to later implement it on
larger datasets. I thus create an input vector x:
* x=rnorm(10)*
I add some noise
* noise=runif(n=10, min=-0.1, max=0.1)*
and I create a simple linear model which calculates my output vector y
* y=2*x+1+noise*

I then do
* my_data <- data.matrix(x)
model = lars(my_data, y, type = 'lasso')*

I then calculate the coefficients (type="coefficients") based on the created
*model **
preds=predict.lars(model)
for(i in 1:10){
est[i]=preds$coef[2]*x[i]
}

y.estimated=est+1+noise

*Then, I apply the same function, predict.lars, but this time with
type="fit".
* preds2=predict.lars(model,my_data)*

When I compare the *y.estimated *to *preds2$fit[,2] *I see that they are not
equal...
I provide you with the returned results:

*y.estimated:*
[2.855597 1.259374 1.673388 1.625999 0.337993 -1.672998
-1.055416 2.423278 4.092116 -1.595545]

*preds2$fit[,2]:*
[2.9120115 1.1790466 1.7452670 1.7239429 0.2893512 -1.6682459
-1.1500982 2.4364527 4.1511509 -1.6098748]

I think they should be equal...Does anyone have an explanation about that?

Thanks a lot for your time!
Eleni C.

[[alternative HTML version deleted]]

==============================================================================
TOPIC: Creating bibtex file of all installed packages?
http://groups.google.com/group/r-help-archive/t/9b384f9c9db3cf07?hl=en
==============================================================================

== 1 of 4 ==
Date: Tues, Dec 15 2009 11:55 pm
From: Achim Zeileis

On Wed, 16 Dec 2009, Romain Francois wrote:

> On 12/16/2009 08:32 AM, Achim Zeileis wrote:
>>
>> On Tue, 15 Dec 2009, Michael Friendly wrote:
>>
>>> Achim and others:
>>>
>>> Achim's solution could be directly usable if it also added a BibTeX key,
>>> perhaps just the name of the package to the '@Manual{,' initial line
>>> of each. I wrapped the previous suggestions in a function, and played
>>> around with the components, but can't quite see how to account for the
>>> failed citation calls. Can anyone take the next step?
>>
>> I had also thought about this after sending my previous solution, so
>> here is an update Michael's version of the function (renamed to the name
>> of the default file). Apart from some smaller touch-ups this has the
>> following changes:
>>
>> - Instead of taking the unique() bibs, I now use the unique() pkgs.
>>
>> (In principle, packages with the same name could be installed in
>> different libraries, potentially containing different citations.
>> But I thought it would be overkill to check for that.)
>>
>> - Citation keys are simply "pkgname" if there is only a single
>> BibTeX item, and "pkgname1" to "pkgnameN" if there are N BibTeX
>> items.
>>
>> (This does not assure that citation keys are unique, though. If
>> there is a package "foo" with 2 citation entries and another package
>> "foo2" with only a single entry, these could be confused. A workaround
>> would be to use "pkgname1" instead of "pkgname" as the citation key
>> even if there is a single citation only. But I thought that would
>> be less intuitive.)
>>
>>
>> Rpackages.bib <- function(file = "Rpackages.bib", verbose = TRUE)
>> {
>> ## installed packages
>> pkgs <- unique(installed.packages()[,1])
>> bibs <- lapply(pkgs, function(x) try(toBibtex(citation(x))))
>> n.installed <- length(bibs)
>>
>> ## omit failed citation calls
>> ok <- !(sapply(bibs, class) == "try-error")
>> pkgs <- pkgs[ok]
>> bibs <- bibs[ok]
>> n.converted <- sum(ok)
>> ## unify to list of Bibtex
>> bibs <- lapply(bibs, function(x) if(inherits(x, "Bibtex")) list(x) else x)
>>
>> ## add bibtex keys to each entry
>> pkgs <- lapply(seq_along(pkgs), function(i) if(length(bibs[[i]]) > 1)
>> paste(pkgs[i], 1:length(bibs[[i]]), sep = "") else pkgs[i])
>> pkgs <- do.call("c", pkgs)
>> bibs <- do.call("c", bibs)
>> for(i in seq_along(pkgs)) bibs[[i]][1] <-
>> gsub("{,", paste("{", pkgs[i], ",", sep = ""), bibs[[i]][1], fixed = TRUE)
>>
>> ## write everything to a single .bib file
>> writeLines(do.call("c", lapply(bibs, as.character)), file)
>> if(verbose) cat("Converted", n.converted, "of", n.installed,
>> "package citations to BibTeX",
>> "\nResults written to file", file, "\n")
>>
>> ## return Bibtex items invisibly
>> invisible(bibs)
>> }
>>
>> Best,
>> Z
>
> And you can read it back into R with package bibtex:
>
>> require( bibtex )
>> entries <- read.bib( "Rpackages.bib" )
>
> but then although the key is read, it is not displayed, because of
> utils:::toBibtex.citation:
>
>> entries[[1]]
> A BibTeX entry for LaTeX users is
>
> @Manual{,
> title = {ant: Version of ant specific to R},
> author = {Romain Francois},
> year = {2009},
> note = {R package version 0.0-10},
> }
>
>> str( entries[[1]] )
> List of 4
> $ title : chr "ant: Version of ant specific to R"
> $ author:List of 1
> ..$ :List of 2
> .. ..$ name : Named chr [1:3] "Romain" "" "Francois"
> .. .. ..- attr(*, "names")= chr [1:3] "first" "middle" "last"
> .. ..$ email: NULL
> .. ..- attr(*, "class")= chr "person"
> ..- attr(*, "class")= chr "personList"
> $ year : chr "2009"
> $ note : chr "R package version 0.0-10"
> - attr(*, "class")= chr "citation"
> - attr(*, "srcref")=Class 'srcref' atomic [1:6] 1 1 6 1 1 1
> .. ..- attr(*, "srcfile")=Class 'srcfile' <environment: 0x8ed067c>
> - attr(*, "entry")= chr "Manual"
> - attr(*, "key")= chr "ant
>
> Should I submit some patch to allow printing of the key when there is a key
> attribute. Maybe the key could also be used for character based indexing of
> citationList objects.

Maybe not at the moment. Kurt and I have been working on some revisions of
the "citation" code which hopefully get incorporated into R-devel early
next year. I'll try to keep this issue in mind.
Z

> Romain
>
> --
> Romain Francois
> Professional R Enthusiast
> +33(0) 6 28 91 30 30
> http://romainfrancois.blog.free.fr
> |- http://tr.im/HlX9 : new package : bibtex
> |- http://tr.im/Gq7i : ohloh
> `- http://tr.im/FtUu : new package : highlight
>
>
>

== 2 of 4 ==
Date: Wed, Dec 16 2009 1:09 am
From: S Devriese

On 12/16/2009 08:32 AM, Achim Zeileis wrote:
> On Tue, 15 Dec 2009, Michael Friendly wrote:
>
>> Achim and others:
>>
>> Achim's solution could be directly usable if it also added a BibTeX key,
>> perhaps just the name of the package to the '@Manual{,' initial line
>> of each. I wrapped the previous suggestions in a function, and played
>> around with the components, but can't quite see how to account for the
>> failed citation calls. Can anyone take the next step?
>
> I had also thought about this after sending my previous solution, so
> here is an update Michael's version of the function (renamed to the
> name of the default file). Apart from some smaller touch-ups this has
> the following changes:
>
> - Instead of taking the unique() bibs, I now use the unique() pkgs.
>
> (In principle, packages with the same name could be installed in
> different libraries, potentially containing different citations.
> But I thought it would be overkill to check for that.)
>
> - Citation keys are simply "pkgname" if there is only a single
> BibTeX item, and "pkgname1" to "pkgnameN" if there are N BibTeX
> items.
>
> (This does not assure that citation keys are unique, though. If
> there is a package "foo" with 2 citation entries and another package
> "foo2" with only a single entry, these could be confused. A
> workaround
> would be to use "pkgname1" instead of "pkgname" as the citation key
> even if there is a single citation only. But I thought that would
> be less intuitive.)
Another possibility would be to use pkgname_versionnumber (e.g.
lattice1226). This would probably provide uniqueness and if used on
subsequent updates of packages does not break older (latex) documents
where previous versions of packages were used.
Just a thought.

Stephan

== 3 of 4 ==
Date: Wed, Dec 16 2009 1:12 am
From: Achim Zeileis

On Wed, 16 Dec 2009, S Devriese wrote:

> On 12/16/2009 08:32 AM, Achim Zeileis wrote:
>> On Tue, 15 Dec 2009, Michael Friendly wrote:
>>
>>> Achim and others:
>>>
>>> Achim's solution could be directly usable if it also added a BibTeX key,
>>> perhaps just the name of the package to the '@Manual{,' initial line
>>> of each. I wrapped the previous suggestions in a function, and played
>>> around with the components, but can't quite see how to account for the
>>> failed citation calls. Can anyone take the next step?
>>
>> I had also thought about this after sending my previous solution, so
>> here is an update Michael's version of the function (renamed to the
>> name of the default file). Apart from some smaller touch-ups this has
>> the following changes:
>>
>> - Instead of taking the unique() bibs, I now use the unique() pkgs.
>>
>> (In principle, packages with the same name could be installed in
>> different libraries, potentially containing different citations.
>> But I thought it would be overkill to check for that.)
>>
>> - Citation keys are simply "pkgname" if there is only a single
>> BibTeX item, and "pkgname1" to "pkgnameN" if there are N BibTeX
>> items.
>>
>> (This does not assure that citation keys are unique, though. If
>> there is a package "foo" with 2 citation entries and another package
>> "foo2" with only a single entry, these could be confused. A
>> workaround
>> would be to use "pkgname1" instead of "pkgname" as the citation key
>> even if there is a single citation only. But I thought that would
>> be less intuitive.)
> Another possibility would be to use pkgname_versionnumber (e.g.
> lattice1226). This would probably provide uniqueness and if used on
> subsequent updates of packages does not break older (latex) documents
> where previous versions of packages were used.
> Just a thought.

No, it doesn't. As pointed out above, the function currently only calls
citation() once for each package name (i.e., ignores multiple versions if
any). The multiple entries do not stem from multiple versions of the
package, but from multiple citations within a single package. See

citation("sandwich")

for an example.
Z

> Stephan
>
>

== 4 of 4 ==
Date: Wed, Dec 16 2009 1:37 am
From: S Devriese

On 12/16/2009 10:12 AM, Achim Zeileis wrote:
> On Wed, 16 Dec 2009, S Devriese wrote:
>
>> On 12/16/2009 08:32 AM, Achim Zeileis wrote:
>>> On Tue, 15 Dec 2009, Michael Friendly wrote:
>>>
>>>> Achim and others:
>>>>
>>>> Achim's solution could be directly usable if it also added a BibTeX
>>>> key,
>>>> perhaps just the name of the package to the '@Manual{,' initial line
>>>> of each. I wrapped the previous suggestions in a function, and played
>>>> around with the components, but can't quite see how to account for the
>>>> failed citation calls. Can anyone take the next step?
>>>
>>> I had also thought about this after sending my previous solution, so
>>> here is an update Michael's version of the function (renamed to the
>>> name of the default file). Apart from some smaller touch-ups this has
>>> the following changes:
>>>
>>> - Instead of taking the unique() bibs, I now use the unique() pkgs.
>>>
>>> (In principle, packages with the same name could be installed in
>>> different libraries, potentially containing different citations.
>>> But I thought it would be overkill to check for that.)
>>>
>>> - Citation keys are simply "pkgname" if there is only a single
>>> BibTeX item, and "pkgname1" to "pkgnameN" if there are N BibTeX
>>> items.
>>>
>>> (This does not assure that citation keys are unique, though. If
>>> there is a package "foo" with 2 citation entries and another
>>> package
>>> "foo2" with only a single entry, these could be confused. A
>>> workaround
>>> would be to use "pkgname1" instead of "pkgname" as the citation key
>>> even if there is a single citation only. But I thought that would
>>> be less intuitive.)
>> Another possibility would be to use pkgname_versionnumber (e.g.
>> lattice1226). This would probably provide uniqueness and if used on
>> subsequent updates of packages does not break older (latex) documents
>> where previous versions of packages were used.
>> Just a thought.
>
> No, it doesn't. As pointed out above, the function currently only
> calls citation() once for each package name (i.e., ignores multiple
> versions if any). The multiple entries do not stem from multiple
> versions of the package, but from multiple citations within a single
> package. See
>
> citation("sandwich")
>
> for an example.
> Z
>
I'm sorry my thoughts were somewhat confusing in the mail. I did catch
that the current mechanism to avoid multiple entries solves the one
package-multiple citations problem and not a multiple package-one
citiation problem. I was just thinking of a way to alter the function
when used sequentially in time. I know that was not the original
poster's problem, but it's one I and several of my colleagues have
encountered often (and until the function posted here, we used a more
manual way of solving the problem of getting the citations of all
installed (or loaded) packages :-). I know that simply using seperate
.bib files for each latex document partially solves the
sequential-in-time problem, but this is actually a different topic, so I
probably should start a different thread.

Stephan

== 1 of 5 ==
Date: Wed, Dec 16 2009 1:26 am
From: Xiang Wu

Is there a function in R that could find the significant digit of a specific
number? Such as for 3.1415, return '5'?

Thanks in advance.

[[alternative HTML version deleted]]

== 2 of 5 ==
Date: Wed, Dec 16 2009 1:32 am
From: Gustaf Rydevik

On Wed, Dec 16, 2009 at 10:26 AM, Xiang Wu <xiang.isu@gmail.com> wrote:

> Is there a function in R that could find the significant digit of a
> specific
> number? Such as for 3.1415, return '5'?
>
> Thanks in advance.
>
> [[alternative HTML version deleted]]
>
> ______________________________________________
> R-help@r-project.org mailing list
> https://stat.ethz.ch/mailman/listinfo/r-help
> PLEASE do read the posting guide
> http://www.R-project.org/posting-guide.html
> and provide commented, minimal, self-contained, reproducible code.
>

Hi,

x<-pi
substr(as.character(x),6,6)

Regards,

Gustaf
--
Gustaf Rydevik, M.Sci.
tel: +46(0)703 051 451
address:Essingetorget 40,112 66 Stockholm, SE
skype:gustaf_rydevik

[[alternative HTML version deleted]]

== 3 of 5 ==
Date: Wed, Dec 16 2009 1:32 am
From: Romain Francois

On 12/16/2009 10:26 AM, Xiang Wu wrote:
>
> Is there a function in R that could find the significant digit of a specific
> number? Such as for 3.1415, return '5'?
>
> Thanks in advance.

Not sure to understand what you mean, but you can have a look at ?signif

Something like this perhaps:

> foo <- function( x, y) abs( round( 10^y * (x - signif(x,y) ) ) )
> foo( 3.1415, 4 )
[1] 5

--
Romain Francois
Professional R Enthusiast
+33(0) 6 28 91 30 30
http://romainfrancois.blog.free.fr
|- http://tr.im/HlX9 : new package : bibtex
|- http://tr.im/Gq7i : ohloh
`- http://tr.im/FtUu : new package : highlight

== 4 of 5 ==
Date: Wed, Dec 16 2009 1:37 am
From: Linlin Yan

Try this:
> f <- function(x) length(gregexpr("[[:digit:]]", as.character(x))[[1]])
> f(3.14)
[1] 3
> f(3.1415)
[1] 5
> f(3.14159265)
[1] 9

On Wed, Dec 16, 2009 at 1:39 PM, Xiang Wu <xiang.isu@gmail.com> wrote:
> Is there a function in R that could find the significant digit of a specific
> number? Such as for 3.1415, return '5'?
>
> Thanks in advance.

== 5 of 5 ==
Date: Wed, Dec 16 2009 4:31 am
From: Duncan Murdoch

On 16/12/2009 12:39 AM, Xiang Wu wrote:
> Is there a function in R that could find the significant digit of a specific
> number? Such as for 3.1415, return '5'?

The question may not have an unambiguous answer: I would say 1.00 and
1.00000 have a different number of significant digits, but once
converted to numbers they are identical in R.

Duncan Murdoch

== 1 of 3 ==
Date: Wed, Dec 16 2009 1:41 am
From: David Scott

I have a client who uses Stata 11.

Can anyone advise me on ways of transferring data from this version of
Stata to R?

Reading the documentation for read.dta in package foreign it seems it
only allows for versions up to Stata 10. I don't know Stata at all but a
bit of poking around on their website led me to believe Stata 11
couldn't write in Stata 10 format. However Stata 11 can write Excel
files in XML, so presumably .xlsx files.

Any advice, correction of my misconceptions, etc. gladly received.

David Scott

_________________________________________________________________
David Scott Department of Statistics
The University of Auckland, PB 92019
Auckland 1142, NEW ZEALAND
Phone: +64 9 923 5055, or +64 9 373 7599 ext 85055
Email: d.scott@auckland.ac.nz, Fax: +64 9 373 7018

Director of Consulting, Department of Statistics

== 2 of 3 ==
Date: Wed, Dec 16 2009 2:24 am
From: Prof Brian Ripley

Have you tried read.dta?

The URL on the help page for read.dta now says 'Stata 11 help for
dta', but it seems to document the same format (114) as Stata 10 (see
the value section). So in so far as Stata is telling us, it looks
like Stata 10 and 11 use the same format (as Stata 8 and 9 did).

On Wed, 16 Dec 2009, David Scott wrote:

> I have a client who uses Stata 11.
>
> Can anyone advise me on ways of transferring data from this version of Stata
> to R?
>
> Reading the documentation for read.dta in package foreign it seems it only
> allows for versions up to Stata 10. I don't know Stata at all but a bit of
> poking around on their website led me to believe Stata 11 couldn't write in
> Stata 10 format. However Stata 11 can write Excel files in XML, so presumably
> .xlsx files.

I suspect it can write .csv files. See for example
http://www.stata.com/help.cgi?outsheet

> Any advice, correction of my misconceptions, etc. gladly received.
>
> David Scott
>
> _________________________________________________________________
> David Scott Department of Statistics
> The University of Auckland, PB 92019
> Auckland 1142, NEW ZEALAND
> Phone: +64 9 923 5055, or +64 9 373 7599 ext 85055
> Email: d.scott@auckland.ac.nz, Fax: +64 9 373 7018
>
> Director of Consulting, Department of Statistics

--
Brian D. Ripley, ripley@stats.ox.ac.uk
Professor of Applied Statistics, http://www.stats.ox.ac.uk/~ripley/
University of Oxford, Tel: +44 1865 272861 (self)
1 South Parks Road, +44 1865 272866 (PA)
Oxford OX1 3TG, UK Fax: +44 1865 272595

== 3 of 3 ==
Date: Wed, Dec 16 2009 5:13 am
From: "JRG"

On 16 Dec 2009 at 10:24, Prof Brian Ripley wrote:

> Have you tried read.dta?
>
> The URL on the help page for read.dta now says 'Stata 11 help for
> dta', but it seems to document the same format (114) as Stata 10 (see
> the value section). So in so far as Stata is telling us, it looks
> like Stata 10 and 11 use the same format (as Stata 8 and 9 did).
>

That's correct: Stata 10 and Stata 11 use the same file format. If you need a still older format,
the -saveold- command in Version 11 of Stata will save a Version 10/11 file in the format that was
used by Versions 8 and 9 of Stata.

---JRG

> On Wed, 16 Dec 2009, David Scott wrote:
>
> > I have a client who uses Stata 11.
> >
> > Can anyone advise me on ways of transferring data from this version of Stata
> > to R?
> >
> > Reading the documentation for read.dta in package foreign it seems it only
> > allows for versions up to Stata 10. I don't know Stata at all but a bit of
> > poking around on their website led me to believe Stata 11 couldn't write in
> > Stata 10 format. However Stata 11 can write Excel files in XML, so presumably
> > .xlsx files.
>
> I suspect it can write .csv files. See for example
> http://www.stata.com/help.cgi?outsheet
>
>
> > Any advice, correction of my misconceptions, etc. gladly received.
> >
> > David Scott
> >
> > _________________________________________________________________
> > David Scott Department of Statistics
> > The University of Auckland, PB 92019
> > Auckland 1142, NEW ZEALAND
> > Phone: +64 9 923 5055, or +64 9 373 7599 ext 85055
> > Email: d.scott@auckland.ac.nz, Fax: +64 9 373 7018
> >
> > Director of Consulting, Department of Statistics
>
> --
> Brian D. Ripley, ripley@stats.ox.ac.uk
> Professor of Applied Statistics, http://www.stats.ox.ac.uk/~ripley/
> University of Oxford, Tel: +44 1865 272861 (self)
> 1 South Parks Road, +44 1865 272866 (PA)
> Oxford OX1 3TG, UK Fax: +44 1865 272595
>
> ______________________________________________
> R-help@r-project.org mailing list
> https://stat.ethz.ch/mailman/listinfo/r-help
> PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
> and provide commented, minimal, self-contained, reproducible code.

John R. Gleason
Associate Professor

Syracuse University
430 Huntington Hall Voice: 315-443-3107
Syracuse, NY 13244-2340 USA FAX: 315-443-4085

PGP public key at keyservers

==============================================================================
TOPIC: MAPE and RMSE in R
http://groups.google.com/group/r-help-archive/t/f2f1aec2869b94d9?hl=en
==============================================================================

== 1 of 1 ==
Date: Wed, Dec 16 2009 1:48 am
From: ruchita gupta

Hello

After performing NLS estimates for some sigmoid model(like logistic growth
model and Gompertz growth models), how can we get the RMSE(root mean square
error) and MAPE(mean absolute percentage error) in R statistical tool for
comparison between two models

Thanks
Ruchita

[[alternative HTML version deleted]]

==============================================================================
TOPIC: Read dataset in R language.
http://groups.google.com/group/r-help-archive/t/22d705f8846e8bc7?hl=en
==============================================================================

== 1 of 2 ==
Date: Wed, Dec 16 2009 2:26 am
From: Nancy Adam

Hi all,

I tried many things to read any dataset but I receive different error messages every time.
One new message is: vector size specified is too large

Can any one please send any part from any dataset just for testing a regression system and to know where is the problem??

Many thanks,
Nancy
_________________________________________________________________
Keep your friends updated—even when you're not signed in.

[[alternative HTML version deleted]]

== 2 of 2 ==
Date: Wed, Dec 16 2009 3:58 am
From: Petr PIKAL

r-help-bounces@r-project.org napsal dne 16.12.2009 11:26:03:

>
> Hi all,
>
> I tried many things to read any dataset but I receive different error
messages
> every time.
> One new message is: vector size specified is too large
>
> Can any one please send any part from any dataset just for testing a
> regression system and to know where is the problem??

Here is a simple data which you can read by

mydat <- read.delim("barva.txt")

However I looked to you earlier post with your data and they are not well
structured. You would need to fix them before trying to read them into R.
Or you can try to read e.g. your trainingset by

read.table("trainingset.txt", sep=",", fill=T, header=T)

and look if all values are correctly aligned to columns

Regards
Petr

>
> Many thanks,
> Nancy
> _________________________________________________________________
> Keep your friends updated—even when you're not signed in.
>
> [[alternative HTML version deleted]]
>
> ______________________________________________
> R-help@r-project.org mailing list
> https://stat.ethz.ch/mailman/listinfo/r-help
> PLEASE do read the posting guide
http://www.R-project.org/posting-guide.html
> and provide commented, minimal, self-contained, reproducible code.

==============================================================================
TOPIC: expand.grid applied to a matrix
http://groups.google.com/group/r-help-archive/t/b151f95f3277f5c1?hl=en
==============================================================================

== 1 of 1 ==
Date: Wed, Dec 16 2009 2:39 am
From: Jean-Christophe Domenge

Many thanks for both replying and fixing the typo!

On Tue, Dec 15, 2009 at 21:20, Charles C. Berry <cberry@tajo.ucsd.edu>wrote:

> On Tue, 15 Dec 2009, Jean-Christophe Domenge wrote:
>
> Dear R gurus,
>> I'm looking for a way to expand a matrix to a data frame as detailed
>> below:
>> given a Matrix M with attribute dimnames=list(c("a","b"),c("u","v")),
>> return
>> a data frame df.M with
>> df.M$row df.M$col df.M$val
>> "a" "u" M["a","u"]
>> "b" "v" M["b". "v"]
>> "a" "u" M["a", "u"]
>> "b" "v" M["b", "v"]
>>
>>
> Do you really want two copies of diag(M) only or is that a typo??
>
> If a typo, try
>
> df.M <- as.data.frame.table( M )
> colnames( df.M ) <- c('row', 'col', 'val' )
>
> If you want what your example shows, then add this line:
>
> df.M <- df.M[ c( 1, 4, 1, 4 ), ]
>
> HTH,
>
> Chuck
>
>
> expand.grid(M) almost does the job as it flattens M to a vector
>> but then the attribute dimnames gets lost.
>> I can always do merge(expand.grid(dimnames(M)), expand.grid(M), by=0)
>> but I'm wondering if there is a more concise way to do it.
>>
>> Thanks for any help!
>>
>> jc
>>
>> [[alternative HTML version deleted]]
>>
>> ______________________________________________
>> R-help@r-project.org mailing list
>> https://stat.ethz.ch/mailman/listinfo/r-help
>> PLEASE do read the posting guide
>> http://www.R-project.org/posting-guide.html
>> and provide commented, minimal, self-contained, reproducible code.
>>
>>
> Charles C. Berry (858) 534-2098
> Dept of Family/Preventive
> Medicine
> E mailto:cberry@tajo.ucsd.edu UC San Diego
> http://famprevmed.ucsd.edu/faculty/cberry/ La Jolla, San Diego 92093-0901
>
>
>

[[alternative HTML version deleted]]

==============================================================================
TOPIC: Reshape a data set
http://groups.google.com/group/r-help-archive/t/0271d00d7ed53aca?hl=en
==============================================================================

== 1 of 1 ==
Date: Wed, Dec 16 2009 2:57 am
From: Henrique Dallazuanna

Try this also:

reshape(test, direction = 'wide', idvar = 'subject', timevar = 'coder')

On Sun, Dec 13, 2009 at 6:04 PM, dadrivr <dadrivr@gmail.com> wrote:
>
> I am trying to reshape a data set. Could someone please help me with the
> reshape, cast, and melt functions? I am new to R and I have tried reading
> up on how to use the reshape package, but I am very confused. Here is an
> example of what I am trying to do:
> subject coder score time
> [1,] 1 1 20 5
> [2,] 1 2 30 4
> [3,] 2 3 10 10
> [4,] 2 2 5 12
> [5,] 3 2 15 NA
> [6,] 3 1 NA 13
>
> Reshape to:
> subject coder.1 score.1 time.1 coder.2 score.2 time.2
> [1,] 1 1 20 5 2 30 4
> [2,] 2 3 10 10 2 5 12
> [3,] 3 2 15 NA 1 NA 13
>
> Here is the code to setup the original data set that I need to reshape:
> subject <- c(1,1,2,2,3,3)
> coder <- c(1,2,3,2,2,1)
> score <- c(20,30,10,5,15,NA)
> time <- c(5,4,10,12,NA,13)
> mydata <- cbind(subject,coder,score,time)
>
> Here is where I'm not sure:
> library(reshape)
> mydata_melt <- melt(mydata, id="subject", na.rm=TRUE)
> mydata_cast <- cast(mydata, ???
>
> Any help would be greatly appreciated. Thanks so much!
> -Isaac
> --
> View this message in context: http://n4.nabble.com/Reshape-a-data-set-tp963104p963104.html
> Sent from the R help mailing list archive at Nabble.com.
>
> ______________________________________________
> R-help@r-project.org mailing list
> https://stat.ethz.ch/mailman/listinfo/r-help
> PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
> and provide commented, minimal, self-contained, reproducible code.
>

--
Henrique Dallazuanna
Curitiba-Paraná-Brasil
25° 25' 40" S 49° 16' 22" O

==============================================================================
TOPIC: R & very large files
http://groups.google.com/group/r-help-archive/t/26c4541c3a7884b3?hl=en
==============================================================================

== 1 of 5 ==
Date: Wed, Dec 16 2009 2:59 am
From: Albert-Jan Roskam

Hi,

I very recently started using R (as in: last week) and I was wondering if anyone could point me to website(s) with sample code to deal with large datasets (length- and/or breadthwise). I understood that R was never designed to work with datasets larger than, say, a couple of hundred Mb. One way is (as I also read) to let R work in conjunction with SQL. That's one interesting approach I'd like to know more about. But I was also hoping that there also were pure R solutions for working with very large tables (was 'scan' designed for that?). In any case, a standard approach would be desirable.

Thanks in advance.

Cheers!!
Albert-Jan

~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
In the face of ambiguity, refuse the temptation to guess.
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

[[alternative HTML version deleted]]

== 2 of 5 ==
Date: Wed, Dec 16 2009 3:08 am
From: S Devriese

On 12/16/2009 11:59 AM, Albert-Jan Roskam wrote:
> Hi,
>
> I very recently started using R (as in: last week) and I was wondering if anyone could point me to website(s) with sample code to deal with large datasets (length- and/or breadthwise). I understood that R was never designed to work with datasets larger than, say, a couple of hundred Mb. One way is (as I also read) to let R work in conjunction with SQL. That's one interesting approach I'd like to know more about. But I was also hoping that there also were pure R solutions for working with very large tables (was 'scan' designed for that?). In any case, a standard approach would be desirable.
>
> Thanks in advance.
>
> Cheers!!
> Albert-Jan
>
> ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
> In the face of ambiguity, refuse the temptation to guess.
> ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
>

See for example the "Large memory and out-of-memory data" section of the
taskview "High-Performance and Parallel Computing with R"
(pick a mirror on http://cran.r-project.org/mirrors.html then "Task
Views" then "HighPerformanceComputing")

Stephan

== 3 of 5 ==
Date: Wed, Dec 16 2009 3:13 am
From: Søren Højsgaard

The sqldf package may be of help to you.
Regards
Søren

-----Oprindelig meddelelse-----
Fra: r-help-bounces@r-project.org [mailto:r-help-bounces@r-project.org] På vegne af Albert-Jan Roskam
Sendt: 16. december 2009 11:59
Til: r-help@r-project.org
Emne: [R] R & very large files

Cheers!!
Albert-Jan

~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
In the face of ambiguity, refuse the temptation to guess.
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

[[alternative HTML version deleted]]

== 4 of 5 ==
Date: Wed, Dec 16 2009 3:13 am
From: Paul Hiemstra

Albert-Jan Roskam wrote:
> Hi,
>
> I very recently started using R (as in: last week) and I was wondering if anyone could point me to website(s) with sample code to deal with large datasets (length- and/or breadthwise). I understood that R was never designed to work with datasets larger than, say, a couple of hundred Mb. One way is (as I also read) to let R work in conjunction with SQL. That's one interesting approach I'd like to know more about. But I was also hoping that there also were pure R solutions for working with very large tables (was 'scan' designed for that?). In any case, a standard approach would be desirable.
>
Hi Albert-Jan,

If you are faced with enormous datasets, R is, in my opinion, a great
tool. It only takes careful thought how to tackle an analysis if the
data does not fit into your memory all at once. As you mentioned, you
could put your data in a database and extract subsets to do the
analysis, lateron combining the results. You could sparsely sample your
data. But without you specifying exactly what it is you want to do, it
is impossible for us to give you any specific advice. Please review the
posting guide for some hints as to what kind of information you can
provide us with.

cheers,
Paul
>
> Thanks in advance.
>
> Cheers!!
> Albert-Jan
>
> ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
> In the face of ambiguity, refuse the temptation to guess.
> ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
>
>
>
> [[alternative HTML version deleted]]
>
>
> ------------------------------------------------------------------------
>
> ______________________________________________
> R-help@r-project.org mailing list
> https://stat.ethz.ch/mailman/listinfo/r-help
> PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
> and provide commented, minimal, self-contained, reproducible code.
>

--
Drs. Paul Hiemstra
Department of Physical Geography
Faculty of Geosciences
University of Utrecht
Heidelberglaan 2
P.O. Box 80.115
3508 TC Utrecht
Phone: +3130 274 3113 Mon-Tue
Phone: +3130 253 5773 Wed-Fri
http://intamap.geo.uu.nl/~paul

== 5 of 5 ==
Date: Wed, Dec 16 2009 5:36 am
From: Albert-Jan Roskam

Thank you all! I got very useful advice and I saw some nice solutions that even made me drool! ;-)

Cheers!!
Albert-Jan

~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
In the face of ambiguity, refuse the temptation to guess.
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

--- On Wed, 12/16/09, Søren Højsgaard <Soren.Hojsgaard@agrsci.dk> wrote:

From: Søren Højsgaard <Soren.Hojsgaard@agrsci.dk>
Subject: SV: [R] R & very large files
To: "Albert-Jan Roskam" <fomcl@yahoo.com>, "r-help@r-project.org" <r-help@r-project.org>
Date: Wednesday, December 16, 2009, 12:13 PM

The sqldf package may be of help to you.
Regards
Søren

Cheers!!
Albert-Jan

~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
In the face of ambiguity, refuse the temptation to guess.
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

[[alternative HTML version deleted]]

==============================================================================
TOPIC: Read dataset in R language
http://groups.google.com/group/r-help-archive/t/01914ec91390e95f?hl=en
==============================================================================

== 1 of 1 ==
Date: Wed, Dec 16 2009 3:47 am
From: "Millo Giovanni"

Hello.
This is to get you started with data.frames, next time please
- read the posting guide
- see the documentation, especially the builtin "R data import/export"
manual form the help menu

## begin R examples, paste into console ##
data(mtcars) # builtin database
class(mtcars) # what it is
mtcars # print it out...
head(mtcars) # ...better: see first rows
fm <- mpg~hp+wt
summary(lm(fm, data=mtcars)) # my favourite regression: see ?mtcars
save(mtcars, file="mtcars.rda") # save as R binary in current working
directory
load(file="mtcars.rda") # load from R binary file, get directly an R
object
write.table(mtcars, file="mtcars.txt") # write to text file (default is
space-delimited)
mtcars <- read.table(file="mtcars.txt") # read from text file and assign
to an R object

## see help for all these functions as
?data
?read.table
## etc.
## end ##

Data import/export is the nastiest step in any stats program, hold on
;^) ...and start with small examples: your "new" message probably means
you don't have enough RAM (although from your post it is impossible to
really tell).
Giovanni

Giovanni Millo
Research Dept.,
Assicurazioni Generali SpA
Via Machiavelli 4,
34132 Trieste (Italy)
tel. +39 040 671184
fax +39 040 671160

------------------------------

Original Message
Date: Wed, 16 Dec 2009 12:26:03 +0200
From: Nancy Adam <nancyadam84@hotmail.com>
Subject: Re: [R] Read dataset in R language.
To: <ehlers@ucalgary.ca>
Cc: r-help@r-project.org
Message-ID: <SNT122-W4335798ADCADAC8E0F70A0B1870@phx.gbl>
Content-Type: text/plain

Hi all,

I tried many things to read any dataset but I receive different error
messages every time.
One new message is: vector size specified is too large

Can any one please send any part from any dataset just for testing a
regression system and to know where is the problem??

Many thanks,
Nancy

Ai sensi del D.Lgs. 196/2003 si precisa che le informazi...{{dropped:13}}

== 1 of 1 ==
Date: Wed, Dec 16 2009 5:39 am
From: Terry Therneau

--- begin included message ---
Hi,

---- end inclusion -----------

Terry Therneau

==============================================================================

You received this message because you are subscribed to the Google Groups "R-help-archive"
group.

To post to this group, send email to r-help@r-project.org or visit http://groups.google.com/group/r-help-archive?hl=en

To unsubscribe from this group, send email to r-help-archive+unsubscribe@googlegroups.com

To change the way you get mail from this group, visit:
http://groups.google.com/group/r-help-archive/subscribe?hl=en

To report abuse, send email explaining the problem to abuse@googlegroups.com

==============================================================================
Google Groups: http://groups.google.com/?hl=en

r-help-archive - 6 new messages in 5 topics - digest

2009-12-16T00:56:00.001-08:00

R-help-archive
http://groups.google.com/group/r-help-archive?hl=en

r-help-archive@googlegroups.com

Today's topics:

* Creating bibtex file of all installed packages? - 2 messages, 2 authors
http://groups.google.com/group/r-help-archive/t/9b384f9c9db3cf07?hl=en
* rcart - classification and regression trees (CART) - 1 messages, 1 author
http://groups.google.com/group/r-help-archive/t/e14c6769fcfa826a?hl=en
* Reshape a data set - 1 messages, 1 author
http://groups.google.com/group/r-help-archive/t/0271d00d7ed53aca?hl=en
* Odp: comparison of these types is not implemented - 1 messages, 1 author
http://groups.google.com/group/r-help-archive/t/3f9eaf80688beb27?hl=en
* lasso regression coefficients - 1 messages, 1 author
http://groups.google.com/group/r-help-archive/t/934b7123088cc14a?hl=en

== 1 of 2 ==
Date: Tues, Dec 15 2009 11:32 pm
From: Achim Zeileis

On Tue, 15 Dec 2009, Michael Friendly wrote:

> Achim and others:
>
> Achim's solution could be directly usable if it also added a BibTeX key,
> perhaps just the name of the package to the '@Manual{,' initial line of each.
> I wrapped the previous suggestions in a function, and played around with the
> components, but can't quite see how to account for the
> failed citation calls. Can anyone take the next step?

I had also thought about this after sending my previous solution, so here
is an update Michael's version of the function (renamed to the name of the
default file). Apart from some smaller touch-ups this has the following
changes:

- Instead of taking the unique() bibs, I now use the unique() pkgs.

(In principle, packages with the same name could be installed in
different libraries, potentially containing different citations.
But I thought it would be overkill to check for that.)

- Citation keys are simply "pkgname" if there is only a single
BibTeX item, and "pkgname1" to "pkgnameN" if there are N BibTeX
items.

(This does not assure that citation keys are unique, though. If
there is a package "foo" with 2 citation entries and another package
"foo2" with only a single entry, these could be confused. A workaround
would be to use "pkgname1" instead of "pkgname" as the citation key
even if there is a single citation only. But I thought that would
be less intuitive.)

Rpackages.bib <- function(file = "Rpackages.bib", verbose = TRUE)
{
## installed packages
pkgs <- unique(installed.packages()[,1])
bibs <- lapply(pkgs, function(x) try(toBibtex(citation(x))))
n.installed <- length(bibs)

## omit failed citation calls
ok <- !(sapply(bibs, class) == "try-error")
pkgs <- pkgs[ok]
bibs <- bibs[ok]
n.converted <- sum(ok)
## unify to list of Bibtex
bibs <- lapply(bibs, function(x) if(inherits(x, "Bibtex")) list(x) else x)

## add bibtex keys to each entry
pkgs <- lapply(seq_along(pkgs), function(i) if(length(bibs[[i]]) > 1)
paste(pkgs[i], 1:length(bibs[[i]]), sep = "") else pkgs[i])
pkgs <- do.call("c", pkgs)
bibs <- do.call("c", bibs)
for(i in seq_along(pkgs)) bibs[[i]][1] <-
gsub("{,", paste("{", pkgs[i], ",", sep = ""), bibs[[i]][1], fixed = TRUE)

## write everything to a single .bib file
writeLines(do.call("c", lapply(bibs, as.character)), file)
if(verbose) cat("Converted", n.converted, "of", n.installed,
"package citations to BibTeX",
"\nResults written to file", file, "\n")

## return Bibtex items invisibly
invisible(bibs)
}

Best,
Z

== 2 of 2 ==
Date: Tues, Dec 15 2009 11:48 pm
From: Romain Francois

On 12/16/2009 08:32 AM, Achim Zeileis wrote:
>
> On Tue, 15 Dec 2009, Michael Friendly wrote:
>
>> Achim and others:
>>
>> Achim's solution could be directly usable if it also added a BibTeX key,
>> perhaps just the name of the package to the '@Manual{,' initial line
>> of each. I wrapped the previous suggestions in a function, and played
>> around with the components, but can't quite see how to account for the
>> failed citation calls. Can anyone take the next step?
>
> I had also thought about this after sending my previous solution, so
> here is an update Michael's version of the function (renamed to the name
> of the default file). Apart from some smaller touch-ups this has the
> following changes:
>
> - Instead of taking the unique() bibs, I now use the unique() pkgs.
>
> (In principle, packages with the same name could be installed in
> different libraries, potentially containing different citations.
> But I thought it would be overkill to check for that.)
>
> - Citation keys are simply "pkgname" if there is only a single
> BibTeX item, and "pkgname1" to "pkgnameN" if there are N BibTeX
> items.
>
> (This does not assure that citation keys are unique, though. If
> there is a package "foo" with 2 citation entries and another package
> "foo2" with only a single entry, these could be confused. A workaround
> would be to use "pkgname1" instead of "pkgname" as the citation key
> even if there is a single citation only. But I thought that would
> be less intuitive.)
>
>
> Rpackages.bib <- function(file = "Rpackages.bib", verbose = TRUE)
> {
> ## installed packages
> pkgs <- unique(installed.packages()[,1])
> bibs <- lapply(pkgs, function(x) try(toBibtex(citation(x))))
> n.installed <- length(bibs)
>
> ## omit failed citation calls
> ok <- !(sapply(bibs, class) == "try-error")
> pkgs <- pkgs[ok]
> bibs <- bibs[ok]
> n.converted <- sum(ok)
> ## unify to list of Bibtex
> bibs <- lapply(bibs, function(x) if(inherits(x, "Bibtex")) list(x) else x)
>
> ## add bibtex keys to each entry
> pkgs <- lapply(seq_along(pkgs), function(i) if(length(bibs[[i]]) > 1)
> paste(pkgs[i], 1:length(bibs[[i]]), sep = "") else pkgs[i])
> pkgs <- do.call("c", pkgs)
> bibs <- do.call("c", bibs)
> for(i in seq_along(pkgs)) bibs[[i]][1] <-
> gsub("{,", paste("{", pkgs[i], ",", sep = ""), bibs[[i]][1], fixed = TRUE)
>
> ## write everything to a single .bib file
> writeLines(do.call("c", lapply(bibs, as.character)), file)
> if(verbose) cat("Converted", n.converted, "of", n.installed,
> "package citations to BibTeX",
> "\nResults written to file", file, "\n")
>
> ## return Bibtex items invisibly
> invisible(bibs)
> }
>
> Best,
> Z

And you can read it back into R with package bibtex:

> require( bibtex )
> entries <- read.bib( "Rpackages.bib" )

but then although the key is read, it is not displayed, because of
utils:::toBibtex.citation:

> entries[[1]]
A BibTeX entry for LaTeX users is

@Manual{,
title = {ant: Version of ant specific to R},
author = {Romain Francois},
year = {2009},
note = {R package version 0.0-10},
}

> str( entries[[1]] )
List of 4
$ title : chr "ant: Version of ant specific to R"
$ author:List of 1
..$ :List of 2
.. ..$ name : Named chr [1:3] "Romain" "" "Francois"
.. .. ..- attr(*, "names")= chr [1:3] "first" "middle" "last"
.. ..$ email: NULL
.. ..- attr(*, "class")= chr "person"
..- attr(*, "class")= chr "personList"
$ year : chr "2009"
$ note : chr "R package version 0.0-10"
- attr(*, "class")= chr "citation"
- attr(*, "srcref")=Class 'srcref' atomic [1:6] 1 1 6 1 1 1
.. ..- attr(*, "srcfile")=Class 'srcfile' <environment: 0x8ed067c>
- attr(*, "entry")= chr "Manual"
- attr(*, "key")= chr "ant

Should I submit some patch to allow printing of the key when there is a
key attribute. Maybe the key could also be used for character based
indexing of citationList objects.

Romain

== 1 of 1 ==
Date: Wed, Dec 16 2009 12:27 am
From: Katie N

Hi,
I am trying to use CART to find an ideal cut-off value for a simple
diagnostic test (ie when the test score is above x, diagnose the condition).
When I put in the model

fit=rpart(outcome ~ predictor1(TB144), method="class", data=data8)

sometimes it gives me a tree with multiple nodes for the same predictor (see
below for example of tree with 1 or multiple nodes). Is there a way to tell
it to make only 1 node? Or is it safe to assume that the cut-off value on
the primary node is the ideal cut-off?

Thanks!
Katie

http://n4.nabble.com/file/n964970/smartDNA%2BCART%2B-%2BTB144n.jpg

http://n4.nabble.com/file/n964970/smartDNA%2BCART%2B-%2BTB122n.jpg

--
View this message in context: http://n4.nabble.com/rcart-classification-and-regression-trees-CART-tp964970p964970.html
Sent from the R help mailing list archive at Nabble.com.

== 1 of 1 ==
Date: Wed, Dec 16 2009 12:44 am
From: xavier.chardon@free.fr

Hi,

You need an additional id column in your data frame, which represents the replication number within each subject (the "1", "2" in the column headings after reshaping in what you showed).

test <- data.frame(subject=c(1,1,2,2,3,3), coder=c(1,2,3,2,2,1), score=c(20,30,10,5,15,NA), time=c(5,4,10,12, NA,13))
test <- ddply(test, .(subject), transform, replication=seq_along(subject))
test_m <- melt(test, id=c("subject", "replication"))
cast(test_m, subject~variable+replication)

HTH,

Xavier

----- Mail Original -----
De: "dadrivr" <dadrivr@gmail.com>
À: r-help@r-project.org
Envoyé: Dimanche 13 Décembre 2009 21h04:17 GMT +01:00 Amsterdam / Berlin / Berne / Rome / Stockholm / Vienne
Objet: [R] Reshape a data set

I am trying to reshape a data set. Could someone please help me with the
reshape, cast, and melt functions? I am new to R and I have tried reading
up on how to use the reshape package, but I am very confused. Here is an
example of what I am trying to do:
subject coder score time
[1,] 1 1 20 5
[2,] 1 2 30 4
[3,] 2 3 10 10
[4,] 2 2 5 12
[5,] 3 2 15 NA
[6,] 3 1 NA 13

Reshape to:
subject coder.1 score.1 time.1 coder.2 score.2 time.2
[1,] 1 1 20 5 2 30 4
[2,] 2 3 10 10 2 5 12
[3,] 3 2 15 NA 1 NA 13

Here is the code to setup the original data set that I need to reshape:
subject <- c(1,1,2,2,3,3)
coder <- c(1,2,3,2,2,1)
score <- c(20,30,10,5,15,NA)
time <- c(5,4,10,12,NA,13)
mydata <- cbind(subject,coder,score,time)

Here is where I'm not sure:
library(reshape)
mydata_melt <- melt(mydata, id="subject", na.rm=TRUE)
mydata_cast <- cast(mydata, ???

Any help would be greatly appreciated. Thanks so much!
-Isaac
--
View this message in context: http://n4.nabble.com/Reshape-a-data-set-tp963104p963104.html
Sent from the R help mailing list archive at Nabble.com.

==============================================================================
TOPIC: Odp: comparison of these types is not implemented
http://groups.google.com/group/r-help-archive/t/3f9eaf80688beb27?hl=en
==============================================================================

== 1 of 1 ==
Date: Wed, Dec 16 2009 12:46 am
From: Petr PIKAL

you have got some response about how to avoid an error but I am curious
what you really want to achieve.

lll<-as.list(rnorm(15))

cbind(unlist(lll),order(unlist(lll)))
[,1] [,2]
[1,] 0.93344282 13
[2,] 0.63501664 14
[3,] 0.05976446 8
[4,] -0.49181944 9
[5,] 2.08002464 7
[6,] 0.53363526 4
[7,] -0.71654312 3
[8,] -1.29946583 11
[9,] -1.08968642 15
[10,] 1.00504020 6
[11,] 0.37454789 12
[12,] 0.63042834 2
[13,] -2.96375833 1
[14,] -1.33535021 10
[15,] 0.52719651 5

unlist(lll)[order(unlist(lll))]
[1] -2.96375833 -1.33535021 -1.29946583 -1.08968642 -0.71654312
-0.49181944
[7] 0.05976446 0.37454789 0.52719651 0.53363526 0.63042834
0.63501664
[13] 0.93344282 1.00504020 2.08002464

Is this what you want?

Regards
Petr

r-help-bounces@r-project.org napsal dne 15.12.2009 11:29:58:

>
> Hi All,
>
> Can you tell me why I get the error message below? It's driving me
nuts.
>
> Thanks,
> Tom
>
> > r_squared
> [[1]]
> [1] 0.9083936
>
> [[2]]
> [1] 0.8871647
>
> [[3]]
> [1] 0.8193883
>
> [[4]]
> [1] 0.728157
>
> [[5]]
> [1] 0.8849525
>
> [[6]]
> [1] 0.8459416
>
> [[7]]
> [1] 0.6702318
>
> [[8]]
> [1] 0.02997816
>
> [[9]]
> [1] 0.8974268
>
> [[10]]
> [1] 0.881217
>
> [[11]]
> [1] 0.8006688
>
> [[12]]
> [1] 0.7207697
>
> [[13]]
> [1] 0.8703734
>
> [[14]]
> [1] 0.8384346
>
> [[15]]
> [1] 0.6237472
>
> >
> > biggest=c(0,0)
> >
> > for (i in 1:15) {
> +
> + if (r_squared[i]>biggest[1]) biggest=c(r_squared[i],i)}
> Error in r_squared[i] > biggest[1] :
> comparison of these types is not implemented
>
>
>
> --
> View this message in context:
http://n4.nabble.com/comparison-of-these-types-
> is-not-implemented-tp964195p964195.html
> Sent from the R help mailing list archive at Nabble.com.
>
> ______________________________________________
> R-help@r-project.org mailing list
> https://stat.ethz.ch/mailman/listinfo/r-help
> PLEASE do read the posting guide
http://www.R-project.org/posting-guide.html
> and provide commented, minimal, self-contained, reproducible code.

== 1 of 1 ==
Date: Wed, Dec 16 2009 12:55 am
From: Eleni Christodoulou

Dear list,

I then do
* my_data <- data.matrix(x)
model = lars(my_data, y, type = 'lasso')*

I then calculate the coefficients (type="coefficients") based on the created
*model **
preds=predict.lars(model)
for(i in 1:10){
est[i]=preds$coef[2]*x[i]
}

y.estimated=est+1+noise

*Then, I apply the same function, predict.lars, but this time with
type="fit".
* preds2=predict.lars(model,my_data)*

When I compare the *y.estimated *to *preds2$fit[,2] *I see that they are not
equal...
I provide you with the returned results:

*y.estimated:*
[2.855597 1.259374 1.673388 1.625999 0.337993 -1.672998
-1.055416 2.423278 4.092116 -1.595545]

*preds2$fit[,2]:*
[2.9120115 1.1790466 1.7452670 1.7239429 0.2893512 -1.6682459
-1.1500982 2.4364527 4.1511509 -1.6098748]

I think they should be equal...Does anyone have an explanation about that?

Thanks a lot for your time!
Eleni C.

[[alternative HTML version deleted]]

==============================================================================

You received this message because you are subscribed to the Google Groups "R-help-archive"
group.

To post to this group, send email to r-help@r-project.org or visit http://groups.google.com/group/r-help-archive?hl=en

To unsubscribe from this group, send email to r-help-archive+unsubscribe@googlegroups.com

To change the way you get mail from this group, visit:
http://groups.google.com/group/r-help-archive/subscribe?hl=en

To report abuse, send email explaining the problem to abuse@googlegroups.com

==============================================================================
Google Groups: http://groups.google.com/?hl=en