Dear all,
i habe a panel data set. One variable was suspected to be endogenous and this was proven by Davidson and macKinnon test and a hausman test of endogeneity. Therefore I used instrumental variables. Random effects specification was preferred by hausman test.
Now my questions:
1. I used dmexogxt for the Davidson and macKinnon test and xtivreg2 for Anderson and Sargan test of validity of the instruments, but both work only for fixed effects. Are there commands for random effects too? Or can infer from them also for the random effects model?
2. I tested for serial correlation using xtserial for the regression without instruments. Is there a way to test for autocorrelation when instrumental variables are present? Is there a way to combine the xtregar command with instrumental variables?
3. One variable enters as INCOME/PRICE. How can I tell Stata to use Instrumental variables for PRICE and then to divide Income by Price?
Thank you for any comments!
Philio
--
GRATIS für alle GMX-Mitglieder: Die maxdome Movie-FLAT!
Jetzt freischalten unter http://portal.gmx.net/de/go/maxdome01
*
* 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/