This is a multi-part message in MIME format.
--------------33BF6EF4428603C70C65AF19
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Brendan Halpin wrote:
> "Buoy" <[EMAIL PROTECTED]> writes:
>
> > There is no problem to fit both row-effect and column-effect models and also
> > uniform association models by SPSS GENLOG procedure. I've got the problem
> > with Row-and-Column effects model (RC model), I don't know how to fit it
> > using SPSS.
> >
> > The model for a two-way table is
> >
> > LOG(F ) = M + L + L + B*U *V
> > ij i j i j
> >
> > where:
> > L = lambda
> > M = constant
> > B = beta
> > in standard notation and:
> > U V
> > i j
> > are estimated parameters, not fixed constants as in Uniform Association
> > Model.
> >
> > I have a feeling that fitting such models (including a term, which is a
> > multiplication of estimated parameters) is impossible via SPSS.
> > Nevertheless, I hope I'm wrong.
>
> I think you're right. My own teaching notes on the topic
> <http://www.iser.essex.ac.uk/~brendan/CDA/foils-master/node51.html>
> suggest using GLIM or lEM
> (<http://cwis.kub.nl/~fsw_1/mto/mto_snw.htm#software>, MSWin only),
> though I have to say I haven't used such log-multiplicative models
> for a very long time.
>
I concur. However, you still might be able to get around this problem by using
SPSS if you can do a little programming in it or simply follow the algorithm below
and do it by hand. Try this,
1. Pick a set of initial scores for U
2. Treat the model with fixed U scores as column-effect model and get estimates of
V from SPSS
3. Use the V estimates from step 2 as fixed score for V and treat the model with
fixed V scores as row-effect model and then get new estimates of U from SPSS
4. Use the U estimates from step 3 as fixed score for U, and cycle back step 2
until the likelihood or the new estimates do not differ too much from the old ones
Goodman suggested it and it works for my problems, hopefully, it will work for
you, too.
--------------33BF6EF4428603C70C65AF19
Content-Type: text/x-vcard; charset=us-ascii;
name="tzuwei.vcf"
Content-Transfer-Encoding: 7bit
Content-Description: Card for Zuwei
Content-Disposition: attachment;
filename="tzuwei.vcf"
begin:vcard
n:Zheng;Zuwei
tel;fax:886-3-4563171 ext 3199
tel;home:886-3-4563171 ext 3122
tel;work:Math Dept, CYCU
x-mozilla-html:FALSE
adr:;;;;;;
version:2.1
email;internet:[EMAIL PROTECTED]
fn:Zuwei Zheng
end:vcard
--------------33BF6EF4428603C70C65AF19--
===========================================================================
This list is open to everyone. Occasionally, less thoughtful
people send inappropriate messages. Please DO NOT COMPLAIN TO
THE POSTMASTER about these messages because the postmaster has no
way of controlling them, and excessive complaints will result in
termination of the list.
For information about this list, including information about the
problem of inappropriate messages and information about how to
unsubscribe, please see the web page at
http://jse.stat.ncsu.edu/
===========================================================================
