topic conditional logistic regression in Statistical Procedures

conditional logistic regression

wajmsu — Sun, 29 May 2016 20:40:56 GMT

Hi,

I am using a matched dataset to carry out conditional logistic regression. Not all matched case-control sets are the same; some cases have 1 control the other may have 2 or 3. Would the analysis be different from if I had similar number of controls for all cases? If yes, how should I address this issue?

Thanks

Re: conditional logistic regression

JacobSimonsen — Mon, 30 May 2016 15:02:22 GMT

You will not get biased result because of different numbers of control-persons. But of course, as more control person the more statistical power you will have. Just remember to put the variable that defines the matching group into the strata statement in proc logistic (it could forexample be the case-ID-number).