[rkward-devel] Status update

[Thomas Friedrichsmeier]

On Thursday 23 July 2009, meik michalke wrote:
> i have installed R version 2.9.1, eRm version 0.10-2 and ltm version 0.9-1.
> what does your output look like?

I get:

Call:
tpm(data = LSAT)

Coefficients:
        Gussng  Dffclt  Dscrmn
Item 1   0.064  -3.342   0.805
Item 2   0.016  -1.553   0.607
Item 3   0.301   0.330  26.415
Item 4   0.065  -2.034   0.570
Item 5   0.029  -3.583   0.552

Log.Lik: -2468.51

Warning message:
In tpm(data = LSAT) :
  Hessian matrix at convergence is not positive definite; unstable solution.

This is after upgrading to R 2.9.1, and updating all installed R packages to 
the latest available version (but was no different with R 2.9.0 and older 
versions of some packages). Same output when running in a plain R session.

Interestingly, on the same machine, dual-booted into windows, I get your 
results. I admit, that does look like something strange is going on on my 
system. No idea, what may be the cause. I suggest you commit your version of 
the test standard for now, and I'll try to find out what may be the issue on my 
computer, later.

> however, i've re-run the tests several times, of course, and i always get
> the exact same results to the last digit. do your results only differ
> slightly from mine, or from run to run as well? i'd suspect the first and
> hold floating point accuracy of different cpus accountable, or something
> like that.

It looks like you're right, but I've only tried two or three times so far.

Regards
Thomas

-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 197 bytes
Desc: This is a digitally signed message part.
URL: <http://mail.kde.org/pipermail/rkward-devel/attachments/20090724/18a7cd8f/attachment.sig>