st: Panel Data Instrumental Variable Test - 2SLS

[Goh Soo Khoon <skgoh@usm.my>]

Dear Statalist

I have a question on Panel data - instrumental variable test (xtivreg)

In Stata, xtivreg allows one to run RE, BE, FE & FD.

Under RE model, there are three options available:
a) use Baltagi EC2SLS RE estimates
b) use Baltagi-Chan instead of Swamny-Aurora estimator
c) ignore instrumental variables and treat all covariates as exogeneous.

Could anyone explain what is option (c)? Since this is an instrumental vari=
able test, why should Stata allow an option ignore instrumental variables? =
 If we select option c, is that means the the 2SLS collapse to RE only??


Can one select RE model, and at the same time, select option (a), (b) and (=
c)??

I use Stata version 12


I read Baltagi (2008), Econometric Analysis of Panel Data.  But I could not  find any explanation on 2SLS withthout instrumental variables!!

Any comment would be very useful and very much appreciated..

thank you

soo khoon
*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   http://www.stata.com/support/faqs/resources/statalist-faq/
*   http://www.ats.ucla.edu/stat/stata/