Hello. Does anybody have some experience with multiple logit- or probit-models, which include an estimate of a "natural response"? I.e. I would like to fit a dose-response-model taking in acount that maybe some individuals die because of other reasons than the influence of the stimuli. (By multipe logit- or probit-models I mean, that I have more than one explanatory variable) Further, I thougt about the idea of constructing an own link-function g, such that range (inv(g)) = [alpha, 1) where alpha > 0. Is anybody familiar with a procedure like this? Sincerely Ruth Meili -- Ruth Meili <meili at stat.math.ethz.ch> Seminar fuer Statistik, LEO D6 ETH (Federal Inst. Technology) 8092 Zurich SWITZERLAND phone: x-41-1-632-3505 fax: x-41-1-632-1228 http://stat.ethz.ch/~meili/ -.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.- 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 _._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._