Stata The Stata listserver
[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]

st: Clogit Model


From   "Jitian Sheu" <jtsheu@bu.edu>
To   <statalist@hsphsun2.harvard.edu>
Subject   st: Clogit Model
Date   Fri, 4 Apr 2003 01:23:46 -0500

Dear Listers:

I am using Conditional Logit model to estimate a choice model.
The total number of individual is around 20000.
Totally, the number of choice is around 50.
Couple choice characteristics are included(around 10 choice-specific
characteristics variables).
However, in my choice model, I also include 4 individual characteristics
variables.
Since, when performing Clogit model, the right hand side variable usualy is
choice-related.
Therefore, for those 4 individual variables, I interact them with choice
dummies(suggested by Green and Stata's Manual).
(And that also means, it extends the number of right hand side variable to
be more than 200.)

However, Stata needs more than 10 hours for convergence.
In fact, I run my first model and got results after 10 hours.
However, for my second similiar model, upto now(Stata has ran more than 10
hours), Stata is still in the process of iteration!

I am just wondering whether anybody knows a better and time-saving method to
do this in Stata.
Or, Clogit usually takes long long long long time?

Thank you very much

JT



*
*   For searches and help try:
*   http://www.stata.com/support/faqs/res/findit.html
*   http://www.stata.com/support/statalist/faq
*   http://www.ats.ucla.edu/stat/stata/



© Copyright 1996–2014 StataCorp LP   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index