Open neilstats opened 1 year ago
Good catch. I hadn't thought about ymax/ymin when doing the 'trickery' code. In my defense, I can't think of a case where I would want to use ymin or ymax in case-control data. Could you give some context, just for my education? (It's still a bug.)
I found this when working with survival data, with >10 strata, rather than case-control.
Within
concordancefit()
, the "Special trickery for matched case-control data" involves changing they
to create disjoint time intervals, but then theymax
argument (and presumably alsoymin
) do not have the intended effect.For example, in all the following
ymax > max(dfr$ptime)
so expect same result as without settingymax
, but the correct result is returned only whenkeepstrata = TRUE
(and the "special trickery" is not used).