Hi All,
I'm trying to run a conditional logistic regression in R (2.14.0) using
clogit from the survival package. The dataset I have is relatively small
(300 observations) with 25 matched strata- there are roughly 2 controls for
each case, and some strata have multiple case/control groups. When I try to
fit a very simple model with a binary outcome and a single continuous
exposure R seems to freeze for a while, and 30 minutes later I receive the
results. However, when I run the exact same conditional logistic regression
model in STATA 10, the exact same answers are produced in <1 second. (Same
coefficients, LR test resutls, standard errors, etc.). I just tried running
an expanded model with covariates that I'd like to control for, but R has
been unresponsive and I doubt it will resolve itself anytime soon. The
syntax I'm using is:
library(survival)
clogit(binary_out~contin_exp + strata(id), data=data)
Various fixes I tried:
1) Upgrading from my prior install version of R 2.10.0 to 2.14.0, no
resolution
2) Increased the memory size limit on R thinking that that might be the
issue, but similar result
I'm running a 32 bit windows machine, i5 CPU at 3.3 Ghz with 3.24 GB of
ram. Any insights into what is occurring/how to increase the speed of this
process would be greatly appreciated. Thanks in advance.
Sincerely,
Vincent
[[alternative HTML version deleted]]
______________________________________________
R-help@r-project.org mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.
