st: CAPM Estimation in Stata using GMM

[Aniruddha Rajan <aniruddha.rajan@googlemail.com>]

Hi all,

I am attempting to estimate the standard Capital Asset Pricing Model over a
set of portfolios using the GMM technique. I have a few questions:

1) I wish to allow for serial correlation in the error terms, as well as
cross-correlation. Does this require the weighting matrix to be specified
as HAC in the options, or the standard error to be specified as HAC?

2) My moment conditions are exactly identified so my parameter estimates
should be equivalent to those obtained through OLS. I merely have a stack
of seemingly unrelated regression equations (SURE) without any instruments.
Does this make any difference regarding what I should specify within the
GMM command in Stata?

3) If possible, I would also like to fit an ARCH/GARCH model to the error
term. Would this be possible using GMM estimation in Stata?

I have easy access to Stata 11 Intercooled so would ideally like to
accomplish my work within this program. However, if this is best performed
using Stata 12 then I shall attempt to locate machine at the university
that has it.

Many thanks and I appreciate any advice that can be provided.

Kind regards,
Aniruddha
*
*   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/