Dear Stata-users,
For my master thesis I am analyzing a sample of panel data (y = whether a company diversified in year t, x = set of company characteristics). Because of the binary dependent variable I use xtlogit, and now want to apply a Hausman test to see whether fixed or random effects models are appropriate.
xtlogit y x, fe
est store fixed
xtlogit y x, re
est store random
hausman fixed random
In some of the tests this delivers a negative Chi-square with the following note:
"Chi2<0 ==> model fitted on these data fails to meet the asymptotic assumptions of the Hausman test; see suest for a generalized test"
In earlier questions I found in the Statlist archive the options 'sigmamore' or 'sigmaless' serve to solve this problem, however, in combination with the xtlogit command this delivers the following message:
"Estimators do not save e(sigma) or e(rmse), sigma option not allowed"
Is any of you familiar with this problem, and if yes, could you tell me whether there is another approach to test for fixed versus random effects or get a positive Hausman outcome? Or do you think this is a case in which I can interpret a negative Hausman score as evidence of failure to reject the null hypothesis (as suggested by Hausman & McFadden, 1984). Please let me know, your help is greatly appreciated!
Kindest regards,
Iris Hesselink
[email protected]
*
* 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/
