[R-sig-ME] (no subject)

Angelo Franchini Angelo.Franchini at bristol.ac.uk
Mon Aug 16 12:57:31 CEST 2010


I am trying to use the lme function part of the nlme package to perform a
random-effects meta-analysis at arm level, but I am not sure what is the
proper use that I should make of the weights option for this situation.
(By arm level I mean a meta-analysis that considers separately both
treatment arm and control arm outcomes, and not just the contrast between

The command that I am using is:
  out <- lme(o~s+t-1, random=~t-1 | s, weights=(~ se.o^2))

s as study/trial identifier
t as 0/1 for control/treatment arm
o as observed outcome in control or treatment arm
se.o as standard error of that outcome measure

I have looked through the R archives and forum and generally on the
Internet for clear indications on what to do, but have not found any.
Could anyone please help with that?

Many thanks.


More information about the R-sig-mixed-models mailing list