search for: sigmaml

...7,0.67,0.67,0.67,0.67,0.67,0.67,0.67,1.08,0.67,0.67, 0.67,0.51) ) ncomp<-5 model<-plsr(y~ .,data=data,ncomp=ncomp,validation="LOO",x="TRUE",y="TRUE",model="TRUE") y<-model$y y.hat<-model$fitted.value[,,ncomp] residuals<-model$residuals[,,ncomp] sigmaML<-sqrt(mean(residuals^2)) loglik<-sum(dnorm(y,y.hat,sigmaML,log=T)) AIC<--2*loglik+2*(ncomp+1) ___________________________________________________________ 好玩贺卡等你发，邮箱贺卡全新上线！ [[alternative HTML version deleted]]

Hello, I'm trying to use AIC to choose between 2 models with positive, continuous response variables and different error distributions (specifically a Gamma GLM with log link and a normal linear model for log(y)). I understand that in some cases it may not be possible (or necessary) to discriminate between these two distributions. However, for the normal linear model I noticed a discrepancy