home *** CD-ROM | disk | FTP | other *** search
- Path: sparky!uunet!dtix!darwin.sura.net!bogus.sura.net!howland.reston.ans.net!zaphod.mps.ohio-state.edu!moe.ksu.ksu.edu!hobbes.physics.uiowa.edu!news.iastate.edu!news
- From: K1.ARC@isumvs.iastate.edu (Arnie Cowan)
- Newsgroups: sci.math.stat
- Subject: Re: SAS: White Heteroskedasticity-Consistent Standard Errors Wanted
- Message-ID: <C01HE5.LJw@news.iastate.edu>
- Date: 29 Dec 92 21:08:28 GMT
- Sender: news@news.iastate.edu (USENET News System)
- Distribution: usa
- Organization: Iowa State University, Ames IA
- Lines: 29
-
- In article <1992Dec29.102446.25292@mic.ucla.edu>,
- iwelch@agsm.ucla.edu (Ivo Welch) writes:
- >
- >In article <1992Dec28.222415.19246@mic.ucla.edu> iwelch@agsm.ucla.edu (Ivo Welch) writes:
- >>
- >>Is there a routine that produces White Heteroskedasticity
- >>consistent standard errors for regressions in SAS?
- >>
- >Thanks to Shinichi Sakata (and other references), I now know that SAS has
- >added an "ACOV" option to proc reg after about version 6.06. Hallelujah!
- >
- Those people still using SAS 5.18 will be glad to know that the
- option is available in that version also. You specify it on the
- MODEL statement:
-
- MODEL A=B / ACOV;
- >>
- >>PS: Actually, this is a pretty basic correction, used in many
- >>papers in my area. Without something equivalent (e.g. a
- >>jack-knife), SAS regression results are almost unpublishable.
- >
- Yes, but remember that White's procedure controls for an
- unspecified form of heteroscedasticity. If you think you know the
- form of heteroscedasticity, something like weighted least squares
- may be more appropriate.
- --
- Arnie Cowan | k1.arc@isumvs.bitnet
- Assistant Professor of Finance | arnie@iastate.edu
- Iowa State University |
-