--On Thursday, July 18, 2002 2:33 -0400 Scott wrote:
Take a look at the Stata FAQs:
Can you explain Chow tests?
http://www.stata.com/support/faqs/stat/chow3.html
How can I compute the Chow test statistic?
http://www.stata.com/support/faqs/stat/chow.html
Scott Merryman
However pooling the two datasets and running a 'chow test' may be overkill,
in that it will also impose the assumption of a common error variance
across the equations. True, a fully interacted second equation will
reproduce OLS point estimates from separate estimations of the two
equations, but the VCV, which depends on the sigma-squared estimate, will
be different (and thus so will be the inference). This can be dealt with,
but IMHO sureg (if feasible) would be a better way to approach this.
Kit
*
* 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/
