|Date: ||Mon, 25 Jun 2007 22:25:05 -0700|
|Reply-To: ||David L Cassell <davidlcassell@MSN.COM>|
|Sender: ||"SAS(r) Discussion" <SAS-L@LISTSERV.UGA.EDU>|
|From: ||David L Cassell <davidlcassell@MSN.COM>|
|Subject: ||Re: Fama MacBeth Correction|
|Content-Type: ||text/plain; format=flowed|
>I was reading a post in the SAS-L list suggesting Pasquariello's
>modification when using FM regressions in order to overcome its drawbacks.
>Has anybody attempted this? If you have any SAS code or any ideas that
>would help, please post your reply.
But *does* Pasquariello's modification overcome the serious drawbacks
of Fama-MacBeth? I have not seen any evidence that it does. (If you
have citations which show this, I'd be interested in seeing them.)
Instead, why don't you use alternative methods which are *known*
to work? If you have panel data (as is usually what people have when
they resort to FM), then why not use appropriate statistical tools
like PROC TSCSREG and PROC PANEL instead?
David L. Cassell
3115 NW Norwood Pl.
Corvallis OR 97330
PC Magazine’s 2007 editors’ choice for best Web mail—award-winning Windows