Hi! I would like to do Cox regression using the available routines in the survival package BUT I want to use an arbitrary link function, i.e. want to use the model h(t)=h_0(t)r(beta'z) with arbitrary function r, instead of h(t)=h_0(t)exp(beta'z) Grateful for any comment on this, Dragi ----------------------------------------------------------- Dragi Anevski, PhD Mathematical Statistics Chalmers University of Technology G?teborg University SE-412 96 G?teborg Sweden +46 31 7725313, +46 706 740980 -.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.- 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 _._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._
On Fri, 2 Aug 2002, Dragi Anevski wrote:> Hi! > I would like to do Cox regression using the available routines in > the survival package BUT I want to use an arbitrary link function, i.e. > want to use the model > > h(t)=h_0(t)r(beta'z) > > with arbitrary function r, instead of > > h(t)=h_0(t)exp(beta'z) > > Grateful for any comment on this,You're out of luck -- the survival package won't be any help with this. The easiest solution would probably be to work out the partial likelihood and its first derivative and use optim() to maximise it. -thomas -.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.- 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 _._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._