Dear R friends. I´m trying to fit a Logistic Regression using glm( family='binomial'). Here is the model: *model<-glm(f_ocur~altitud+UTM_X+UTM_Y+j_sin+j_cos+temp_res+pp, offset=(log(1/off)), data=mydata, family='binomial')*
mydata has 76820 observations. The response variable f_ocur) is a 0-1. This data is a SAMPLE of a bigger dataset, so the idea of setting the offset is to account that the data used here represents a sample of the real data to be analyzed. For some reason the offset is not working. When I run this model I get a result, but when I run the same model but without the offset I get the exact result than the previous model I was expecting a different result but no... there is no difference. Am I doing something wrong? Should the offset be with the linear predictors? like this: *model<-glm(f_ocur~altitud+UTM_X+UTM_Y+j_sin+j_cos+temp_res+pp+** offset(log(1/off))**, data=mydata, family='binomial')* Once the model is ready, I´d like to use it in a new data. The new data would be the data to validate this model, this data has de the same columns, my idea is to use: *validate<-predict(model, newdata=data2, type='response')* And here comes my question, does the predict function takes into consideration the *offset *used to create the model? if not, what should I do in order to get the correct probabilities in the new data? I´d really appreciate if anyone could help me. Thank you. Lucas. [[alternative HTML version deleted]]
______________________________________________ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.