carol white wrote:> Hi,
> How is the combined value (displayed by a diamond on the forest plot) of a
meta-analysis illustrated by a forest plot calculated? The applied method and
all calculation details are welcome.
>
> Thanks
>
> Carol
>
This is not a very well posed question. First of all there are two ways
of getting a diamond in a forest plot, using either fixed or random
effects meta analysis. Secondly there are at least two R packages which
do meta-analysis, meta and rmeta. To which are you referring? They may
do something a little different.
Despite this you can find the answer yourself by looking at the code.
For example in the package rmeta, you probably need to look at the code
for meta.DSl and meta.MH for the calculation of (respectively) random
and fixed meta-analyses.
If you can't see what the code is doing and perhaps don't know the
algorithms typically used in meta-analysis, there is a nice summary of
the methods used in the revman package from the Cochrane collaboration:
http://www.cc-ims.net/revman/documentation/Statistical-methods-in-RevMan-5.pdf
Overall your question has the flavour of asking other people to just
give you the answer without any work on your behalf. You would be
advised to read the posting guide as recommended on the bottom of each
R-help message, otherwise you might receive replies which appear to be
quite offensive.
David Scott
--
_________________________________________________________________
David Scott Department of Statistics
The University of Auckland, PB 92019
Auckland 1142, NEW ZEALAND
Phone: +64 9 923 5055, or +64 9 373 7599 ext 85055
Email: d.scott at auckland.ac.nz, Fax: +64 9 373 7018
Director of Consulting, Department of Statistics