[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]
Re: st: Displaying system variables after a meta-analysis using 'meta'
At 14:01 10/05/2005, Roger Harbord wrote:
Another possible package for doing a meta-analysis is -metaparm-,
downloadable from SSC using the -ssc- command. Whether you use -metan- or
-metaparm- to do the calculations, you might like to use the -eclplot-
package (also downloadable from SSC) to do the confidence interval plots,
as -eclplot- is written in Stata 8, which contains better graphics than
previous Stata versions.
--On 09 May 2005 17:43 +0100 Andrea Mann <email@example.com> wrote:
I have performed a meta-analysis using the 'meta' command and would like
to outfile the odds ratio, lower confidence limit, upper confidence limit
and fixed and random weights, for each individual study, to a raw file.
What are the names of the system variables for the odds ratio, lower ci,
upper ci and the weights? I intend to generate variables from these
system variables, collapse, and outfile. Is that correct?
It's probably going to be a little easier if you switch to use -metan-
instead of -meta- as -metan- generates new variables containing the
weights and CI limits. The most recent version of -metan- accepts the same
input variables in the same order as -meta- i.e. effect estimates and
standard errors. This is version 1.81 and is obtained by typing -ssc
intall metaaggr, replace-. Full documentation including names of the
variables left behind is available at
How you best handle the variables afterwards depends on what you want to
use them for which you know better than us.
Hope this helps.
Lecturer in Medical Statistics
Department of Public Health Sciences
Division of Asthma, Allergy and Lung Biology
King's College London
5th Floor, Capital House
42 Weston Street
London SE1 3QD
Tel: 020 7848 6648 International +44 20 7848 6648
Fax: 020 7848 6620 International +44 20 7848 6620
or 020 7848 6605 International +44 20 7848 6605
Opinions expressed are those of the author, not the institution.
* For searches and help try: