Dr. Tweedie has a few papers on this topic with three (former) students of
his and others, the latest with Dr. Sue Duval are to appear in Biometrics
and JASA. The Stat. in Med. paper of his on meta-analysis is with me (also
a former student of his), but it's on estimation of the heterogeneity
variance and doesn't account at all for publication bias.
See his home page
http://www.biostat.umn.edu/~tweedie/documents/tweediecurrentpapers.html for
the publication bias papers (see trim1.ps/trim1.pdf and trim2.ps/trim2.pdf
for the Biometrics and JASA papers) and our Stat in Med paper
(Finaltau2.ps), or mine http://www.stat.colostate.edu/~bradb for the Stat in
Med paper.
I have written fairly complete software in S-Plus to do almost all of this,
including most of my variance estimation work and Duval and Tweedie's
publication bias methodology. When I've gotten it a bit more checked out, I
might post to statlib. Note that these methods are for one-way models,
where there are no covariates. The recent paper by Thompson and Sharp in
Stat. in Med., for example, should be consulted in these cases.
Regards,
Brad
Brad Biggerstaff, Ph.D.
Mathematical Statistican
Centers for Disease Control and Prevention
National Center for Infectious Diseases
Division of Vector-Borne Infectious Diseases
P.O. Box 2087
Fort Collins, CO 80522-2087 U.S.A.
(970) 221-6473 bkb5@cdc.gov
> -----Original Message-----
> From: In-Sun Nam [SMTP:i.nam@fsc.qut.edu.au]
> Sent: Monday, December 20, 1999 6:56 PM
> To: S-List
> Subject: [S] reference
>
>
> Dear all,
>
> Does anyone know a reference or two regarding a test for the existence
> and number of missing studies (for Meta analysis), possibly by Tweedie,
> possibly in Stats. in Med. or in Stat. Society??
>
> Thanks in advance,
> In-Sun
>
> -----------------------------------------------------------------------
> This message was distributed by s-news@wubios.wustl.edu. To unsubscribe
> send e-mail to s-news-request@wubios.wustl.edu with the BODY of the
> message: unsubscribe s-news
-----------------------------------------------------------------------
This message was distributed by s-news@wubios.wustl.edu. To unsubscribe
send e-mail to s-news-request@wubios.wustl.edu with the BODY of the
message: unsubscribe s-news
|