GLMM in SPSS 19 help

[cfox]

Hi

I've just installed a trial version SPSS 19 for Mac and am having immediate problems.

-the program is extremely slow, even with 4GB memory.

-After trying to run a multinomial GLMM with repeated measures, SPSS output states that I cannot run a multinomial logistic with repeated measures. Is this an error or can SPSS 19 simply not run such a model? If the latter, what program might be useful (I don't know R but do know SAS)?

-Trying to circumvent the above problem, I ran a binomial GLMM with repeated measures and receive a notice that the "processor is unavailable". Again, is this a program glitch? Any fixes?

Thanks in advance,

CF.

[Ryan]

CF,

I doubt SPSS 19 is capable of handling a repeated measures multinomial
logistic regression. SAS can certainly handle such a model. Here's an
example [with data simulation code] I posted to SAS-L a while back:

http://www.listserv.uga.edu/cgi-bin/wa?A2=ind1102A&L=sas-l&P=R6248

Note that I assumed the correlation across random effects was zero. If
you want to allow for correlated random effects, you will have to use
the procedure NLMIXED and change the RANDOM statement provided in the
post linked above.

HTH,

Ryan

On Thu, Jul 14, 2011 at 4:03 PM, cfox <[hidden email]> wrote:

> Hi
>
> I've just installed a trial version SPSS 19 for Mac and am having immediate
> problems.
>
> -the program is extremely slow, even with 4GB memory.
>
> -After trying to run a multinomial GLMM with repeated measures, SPSS output
> states that I cannot run a multinomial logistic with repeated measures. Is
> this an error or can SPSS 19 simply not run such a model? If the latter,
> what program might be useful (I don't know R but do know SAS)?
>
> -Trying to circumvent the above problem, I ran a binomial GLMM with repeated
> measures and receive a notice that the "processor is unavailable". Again, is
> this a program glitch? Any fixes?
>
> Thanks in advance,
>
> CF.
>
> --
> View this message in context: http://spssx-discussion.1045642.n5.nabble.com/GLMM-in-SPSS-19-help-tp4587971p4587971.html
> Sent from the SPSSX Discussion mailing list archive at Nabble.com.
>
> =====================
> To manage your subscription to SPSSX-L, send a message to
> [hidden email] (not to SPSSX-L), with no body text except the
> command. To leave the list, send the command
> SIGNOFF SPSSX-L
> For a list of commands to manage subscriptions, send the command
> INFO REFCARD
>

=====================
To manage your subscription to SPSSX-L, send a message to
[hidden email] (not to SPSSX-L), with no body text except the
command. To leave the list, send the command
SIGNOFF SPSSX-L
For a list of commands to manage subscriptions, send the command
INFO REFCARD