R help - Nov 2012 - Function storing error messages in 32 bit R-2.15.2 version

I am using 32 bit R - 2.15.2 version and working on package ltm. The
program does not give me any warning message when I run tpm command for a
dataset. If run tpm command for the same dataset in 64-bit R 2.15.1
version, I get the warning message shown below.

Can someone suggest why is no warning message being generated by 32-bit R?
Or is it being generated but not getting printed? Which function will have
this warning message stored?

*Consider the following code that I ran on 64-bit R - *

This is package 'ltm' version '0.9-7'
> resp<-read.table("D:\\DriveE\\IRT\\Outfiles\\testFile.txt")
> params<-tpm(resp)
Warning message:
*In tpm(resp) :*
*  Hessian matrix at convergence contains infinite or missing values;
unstable solution.*

*Consider the following code that I ran on 32-bit R - *
> resp<-read.table("D:\\DriveE\\IRT\\Outfiles\\testFile.txt")
> params<-tpm(resp)
>warnings()
NULL

Thanks and regards,
Maulik Shah

	[[alternative HTML version deleted]]

I think we need the data in order to see if this is a bug or "just"
some
numerical issue.

Uwe Ligges



On 21.11.2012 07:14, Maulik Shah wrote:
> I am using 32 bit R - 2.15.2 version and working on package ltm. The
> program does not give me any warning message when I run tpm command for a
> dataset. If run tpm command for the same dataset in 64-bit R 2.15.1
> version, I get the warning message shown below.
>
> Can someone suggest why is no warning message being generated by 32-bit R?
> Or is it being generated but not getting printed? Which function will have
> this warning message stored?
>
> *Consider the following code that I ran on 64-bit R - *
>
> This is package 'ltm' version '0.9-7'
>
>>
resp<-read.table("D:\\DriveE\\IRT\\Outfiles\\testFile.txt")
>> params<-tpm(resp)
> Warning message:
> *In tpm(resp) :*
> *  Hessian matrix at convergence contains infinite or missing values;
> unstable solution.*
>
> *Consider the following code that I ran on 32-bit R - *
>
>>
resp<-read.table("D:\\DriveE\\IRT\\Outfiles\\testFile.txt")
>> params<-tpm(resp)
>> warnings()
> NULL
>
> Thanks and regards,
> Maulik Shah
>
> 	[[alternative HTML version deleted]]
>
> ______________________________________________
> R-help at 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 data we used for testing is pasted below - 
<http://r.789695.n4.nabble.com/file/n4650532/testfile.png> 

Note that our objective is to analyse a dataset with 60 items and about
25,000 candidates but we used the above to test if we get desired result and
the two versions gave different results.



--
View this message in context:
http://r.789695.n4.nabble.com/Function-storing-error-messages-in-32-bit-R-2-15-2-version-tp4650254p4650532.html
Sent from the R help mailing list archive at Nabble.com.

I am working with Maulik on this. We installed a 64-bit version of R 2.15.1
now but the ltm package is under version 2.15.2
The same problem is occurring even with this version.



--
View this message in context:
http://r.789695.n4.nabble.com/Function-storing-error-messages-in-32-bit-R-2-15-2-version-tp4650254p4650534.html
Sent from the R help mailing list archive at Nabble.com.

On Nov 23, 2012, at 2:21 AM, Paarkhi wrote:
> I am working with Maulik on this. We installed a 64-bit version of R  
> 2.15.1
> now but the ltm package is under version 2.15.2
> The same problem is occurring even with this version.
>
You and Maulik are posting messages without context and without the  
means to reproduce the behavior you are concerned about. You are also  
asking a specific question about a package that is not widely used.  
The Posting Guide suggests that you always include context, always  
make examples reproducible, and it also suggests that questions about  
aberrant package behavior should go to the maintainer than Rhelp.
>
> --
> View this message in context:
http://r.789695.n4.nabble.com/Function-storing-error-messages-in-32-bit-R-2-15-2-version-tp4650254p4650534.html
> Sent from the R help mailing list archive at Nabble.com.
>
No. It was sent to Nabble which is not Rhelp or the Rhelp Archives.

-- 

David Winsemius, MD
Alameda, CA, USA