Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at statalist.org.
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
st: vce(robust) and suest based Hausman
From
Ioannis Tikoudis <[email protected]>
To
"'[email protected]'" <[email protected]>
Subject
st: vce(robust) and suest based Hausman
Date
Mon, 11 Jul 2011 16:16:57 +0200
Dear Statalist,
I have the following questions on mlogit:
1) It seems that is possible to estimate an MNL model and get an
adjusted covariance matrix for the MNL estimators. While this is
straightforward in OLS (using the residual covariance matrix), I do not
follow how this is done in STATA. So does it replace the standard MNL
choice probabilities in the last iteration of the likelihood
maximization, with other choice probabilities?
2) what is the exact difference between the McFadden Hausman IIA test
and the suest based Hausman test? What is the test statistic of the last
one and how is it performed? Thank you very much in advance
--
Ioannis Tikoudis
(M.Phil, Tek.Lic)
Division of Economics
The Royal Institute of Technology
Drottning Kristinas väg 30
SE-100 44 Stockholm
Phone: +46 (0) 87906969
Cell: +46 (0) 739193271
*
* 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/