[R] exact confidence intervals for conditional logistic regression

vito muggeo vito.muggeo at giustizia.it
Wed Feb 20 09:29:23 CET 2002


Hi,
I *suppose* that exact logistic regression is not (yet?!!?) implemented in
any R software, because it should be very hard to calculate the
permutational distribution;
Possible alternative (beside buying LogXact) are using MCMC methods (maybe
using (Win)BUGS, but I am not able to tell you anything more) or using
double saddlepoint approximation to approximate the *univariate* conditional
distribution: the S-Plus glm.cond() by A.Brazzale does this (I don't whether
it works in R, too). However  recent papers by Corcoran et al. (Stat in Med,
2001, pg. 2723) shows that this approach, sometimes produces unreliable
results.
Hope that this helps you,
vito

Please let me know if you know something more about computational aspects



----- Original Message -----
From: "Tomas Aragon" <Tomas.Aragon at sfdph.org>
To: <r-help at stat.math.ethz.ch>
Cc: <wayne.enanoria at sfdph.org>
Sent: Tuesday, February 19, 2002 11:03 PM
Subject: [R] exact confidence intervals for conditional logistic regression


> Dear R folks,
> We completed a matched case-control study that was analyzed using
> conditional logistic regression. Because of the small sample size we need
> to calculate exact confidence intervals. The quick solution is to purchase
> LogExact by Cytel. However, we'd like to do this in R. Anyone have
> experience with this?
>
> Many thanks,
> Tomas
> ____________________________________
> Tomas Aragon, MD, DrPH, Director
> Com Health Epidemiology & Disease Control
> San Francisco Department of Public Health
> Tel: (415) 554-2724, Fax: (415) 554-2848
> Pgr: (415) 809-7837, Mob: (415) 706-5203
> Email: tomas.aragon at sfdph.org
> URL: http://www.medepi.org/aragon
>
> -.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.
-.-.-
> r-help mailing list -- Read
http://www.ci.tuwien.ac.at/~hornik/R/R-FAQ.html
> Send "info", "help", or "[un]subscribe"
> (in the "body", not the subject !)  To: r-help-request at stat.math.ethz.ch
>
_._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._.
_._
-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-
r-help mailing list -- Read http://www.ci.tuwien.ac.at/~hornik/R/R-FAQ.html
Send "info", "help", or "[un]subscribe"
(in the "body", not the subject !)  To: r-help-request at stat.math.ethz.ch
_._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._



More information about the R-help mailing list