Dear all,
I have to analyse survival using Cox regression. There is some missing
covariates in my data set and I would like to perform multiple imputation.
The MICE package doesn't allows to combine multiple imputation for a Cox
model.
Is there an SPLUS/R code to do this?
Regards,
Roch
Hi,
In R you can do the imputation part using mice and then fit the models
and pool them using the package mitools. mitools has a nice vignette
that explains how to use it.
R> require(mitools)
R> vignette("smi")
R> # (almost) from ?coxph
R> test2 <- data.frame(
start=c(1, 2, 5, 2, 1, 7, 3, 4, 8, 8),
stop =c(2, 3, 6, 7, 8, 9, 9, 9,14,17),
event=c(1, 1, 1, 1, 1, 1, 1, 0, 0, 0),
x =c(1, 0, NA, 1, NA, 1, 1, 1, 0, 0) )
# using mice to impute
R> mi <- mice( test2 )
# prepare the data for mitools
R> mit <- imputationList( lapply( 1:5, complete, x = mi ) )
# run the models
R> models <- with( mit, coxph( Surv(start, stop, event)~x ) )
# combine them
R> MIcombine( models )
Multiple imputation results:
with.imputationList(mit, coxph(Surv(start, stop, event) ~ x))
MIcombine.default(models)
results se
x -0.4694952 0.9454943
R> summary( MIcombine( models ) )
Multiple imputation results:
with.imputationList(mit, coxph(Surv(start, stop, event) ~ x))
MIcombine.default(models)
results se (lower upper) missInfo
x -0.4694952 0.9454943 -2.328277 1.389286 10 %
Cheers,
Romain
--
Mango Solutions
Tel +44 1249 467 467
Fax +44 1249 467 468
Mob +44 7813 526 123
data analysis that delivers
_____________________________________________________
Roch GIORGI, MD, PhD
Service de Santé Publique et d'Information Médicale
Hôpital de la Timone
264, rue St Pierre
13385 Marseille cedex 5
Tel: +33 (0)491 384 949
Fax: +33 (0)491 385 749
email: roch.giorgi@ap-hm.fr
Web: http://cybertim.timone.univ-mrs.fr/
--------------------------------------------------------------------
This message was distributed by s-news@lists.biostat.wustl.edu. To
unsubscribe send e-mail to s-news-request@lists.biostat.wustl.edu with
the BODY of the message: unsubscribe s-news
|