st: Heteroscedastic logit model

[Dieter Somers <dietersomers@gmail.com>]

Hi everyone,

I am trying to use a heteroscedastic version of the conditional logit model.
I'm using the clogithet command. However, I am not able to run the regression.
Next from the option 'group( )' which identifies the groups, you also
have to specify the heteroscedastic variables with the 'het()'.
However, when I run the regression, I get an error that states that
the variables that cause the heteroscedasticity are not constant
within the groups.
This seems strange, because in a conditional logit model, these
variables are obviously different from each other within each group,
or am I missing something?
I was also wondering whether there exists a general test that checks
for heteroscedasticity in a conditional logit model (for instance a
command similar to the 'hettest') ?


Kind regards,

Dieter
*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   http://www.stata.com/support/statalist/faq
*   http://www.ats.ucla.edu/stat/stata/