Hi,
I estimate a -mlogit-model with robust (Huber-White) standard errors. When
using the user-command -mlogtest- I get the error message (both in Stata 7
and 8) to use the -force- option.
--------begin---
. mlogit lu00 $list00, basecategory(`basecat') nolog robust
. mlogtest, lr lrcomb iia
**** Likelihood-ratio tests for independent variables
Ho: All coefficients associated with given variable(s) are 0.
lu00 | chi2 df P>chi2
-------------+-------------------------
lrtest not valid after robust
specify force option to perform test anyway
r(198);
--------end---
The help for -lrtest- tells that:
--------begin---
force suppresses the tests for error conditions that potentially invalidate
the use of
likelihood ratio tests. The following conditions are normally verified:
- the models were not fitted with robust, cluster()
--------end---
(Clustering brought up the same error message.) What does the last sentence
of the help imply? Do robust standard errors invalidate -lrtest-s? Why so?
Thanks in advance, greetings,
--- Daniel ---
[email protected]
*
* 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/
