Dear Stata community,
My question concerns the order of testing for mis-specification in STATA
I have some panel data (31 countries and 6- four year periods). I
ususpect that two of the regressors are endogous and I want to use the
Hausman test to test for endogeneity one after the other.
I am fairly sure my data is neither heteroskedastic nor
auto-correlated but I want to test for these.
Should I test for endogeneity (correcting for heteroskedasticity and
auto-correlation)
What STATA commands will I use?
Or
Should I first test for heteroskedasticity and auto-correlation, then
based on what I find test for endogeneity?
*
* 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/
