Re: st: Heteroskedasticity logit model

[Richard Williams <richardwilliams.ndu@gmail.com>]

The user-written -oglm- routine available at SSC can estimate ordinal 
and binary hetero models using logit, probit, and other links. There 
are various controversies concerning such models. See, for example,

http://www3.nd.edu/~rwilliam/xsoc73994/RW_ESRA2013.pptx

and the sources it cites. The oglm support page is at

http://www3.nd.edu/~rwilliam/oglm/index.html

Incidentally, I would probably just go ahead and use hetprob unless 
you really need the other features that oglm offers.

At 06:47 AM 12/11/2013, Nerissa Fanoembi wrote:
> Hello,
> I am running a pooled logit model. The data is randomly sampled 
> cross sections from 2004 till 2012. I want to check the 
> heteroskedasticity problem. But there is no heteroskedasticity test 
> available for the logit model. How is it possible to test 
> heteroskedasticity then? The probit model is almost similar as the 
> probit model. Therefore I run the hetprob command where I include 
> all independent variables (categorical and continuous independent 
> variables). The test shows that heteroscedasticity is a problem (the 
> prob of the Likelihood-ratio test of lnsigma2 is 0.0001). In 
> addition, what can I do to solve the heteroskedasticity problem in 
> the logit model? And what is the useful procedure to test 
> heteroskedasticity in a logit model?
> Best regards,
> Nerissa Fanoembi.
> *
> *   For searches and help try:
> *   http://www.stata.com/help.cgi?search
> *   http://www.stata.com/support/faqs/resources/statalist-faq/
> *   http://www.ats.ucla.edu/stat/stata/

-------------------------------------------
Richard Williams, Notre Dame Dept of Sociology
OFFICE: (574)631-6668, (574)631-6463
HOME:   (574)289-5227
EMAIL:  Richard.A.Williams.5@ND.Edu
WWW:    http://www.nd.edu/~rwilliam

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