st: Average Rolling Regression Coefficient Significance

["Alexander v. Angerer" <avangerer@web.de>]

Hello,

I have a question regarding the rollreg and rolling routine.

Let's simplify that I have two portfolios. My hypothesis is  that one portfolio has a higher CAPM beta than the other portfolio when performing a rolling regression.

Let's assume i take the mean of the Beta of the individual regressions. Is there any test that can prove significance of the AVERAGE beta coefficient of the rolling regression?

When testing the hypothesis of a higerer mean beta for one of the two portfolios I assume due to the overlapping window periods when calculating the individual betas, I cannot perform a normal t-test for for difference in mean (in betas) with unequal variance for the two portfolios? Is there another useful test?

Best regards,

Alexander

___________________________________________________________
Empfehlen Sie WEB.DE DSL Ihren Freunden und Bekannten und wir   
belohnen Sie mit bis zu 50,- Euro! https://freundschaftswerbung.web.de

*
*   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/