MCMCglmm multivariate meta-analysis with covariance

2 messages · Jon Bischof, Viechtbauer Wolfgang (STAT)

Tue, Oct 18, 2016 6:24 PM #

I'm interested in fitting a multivariate meta-analysis model with
correlated measurement error. This means fixing the error to a covariance
matrix per row.

I saw this post
<https://stat.ethz.ch/pipermail/r-sig-mixed-models/2013q2/020180.html> on
the mailing list about non-correlated outcomes, but the noise correlation
is too large to ignore in my use case. Professor Hadfield implies in the
post that it is possible but "complicated". Does anyone know how to do it?

Thanks!
Jon Bischof

Viechtbauer Wolfgang (STAT)

Wed, Oct 19, 2016 12:50 AM #

For ML/REML estimation, you can also use metafor, mvmeta, and metaSEM.

An illustration of a multivariate meta-analysis with the metafor package can be found here:

http://www.metafor-project.org/doku.php/analyses:berkey1998

Best,
Wolfgang

Wolfgang Viechtbauer, Ph.D., Statistician | Department of Psychiatry and    
Neuropsychology | Maastricht University | P.O. Box 616 (VIJV1) | 6200 MD    
Maastricht, The Netherlands | +31 (43) 388-4170 | http://www.wvbauer.com    

> -----Original Message-----
> From: R-sig-mixed-models [mailto:r-sig-mixed-models-bounces at r-
> project.org] On Behalf Of Jon Bischof
> Sent: Wednesday, October 19, 2016 03:25
> To: r-sig-mixed-models at r-project.org
> Subject: [R-sig-ME] MCMCglmm multivariate meta-analysis with covariance
> 
> I'm interested in fitting a multivariate meta-analysis model with
> correlated measurement error. This means fixing the error to a covariance
> matrix per row.
> 
> I saw this post
> <https://stat.ethz.ch/pipermail/r-sig-mixed-models/2013q2/020180.html> on
> the mailing list about non-correlated outcomes, but the noise correlation
> is too large to ignore in my use case. Professor Hadfield implies in the
> post that it is possible but "complicated". Does anyone know how to do
> it?
> 
> Thanks!
> Jon Bischof