Bookmark and Share

Notice: On March 31, it was announced that Statalist is moving from an email list to a forum. The old list will shut down on April 23, and its replacement, statalist.org is already up and running.


[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

st: Estimation of consistent var-covariance matrix of probit estimates


From   Fabio Zona <fabio.zona@unibocconi.it>
To   statalist@hsphsun2.harvard.edu
Subject   st: Estimation of consistent var-covariance matrix of probit estimates
Date   Tue, 23 Feb 2010 01:08:44 +0100 (CET)

Dear all,

I am dealing with a cross-section, probit/logit regression. The unit of analysis is the dyad (pair of firms); in these cases, regressions suffer from strong autocorrelation.

In this context, a particular procedure, named the QAP (Quadratic Assignment Procedure) allows to correct for autocorrelation and to get correct standard errors.
However, QAP does NOT calculate any variance-covariance matrix of estimates: I do need this matrix...

My question for you:
- if I run a probit/logit (without the _qap correction for standard errors) and, hence, the regression suffers from strong autocorrelation, is the variance-covariance matrix biased? (I think so, please, confirm this!)

- is there a tool in STATA to get a correct var-cov matrix of estimates when a probit/logit model is affected by strong autocorrelation due to dyadic data?

Thank you very much
*
*   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/


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