I have fitted a logistic regression model in r using 'glm' function. It returns two warning messages that "algorithm did not converge " and "fitted probabilities numerically 0 or 1 occurred". How can I resolve these problems?, More specifically, the data are ym<-c(0, 0, 0, 1, 1, 0 ,1, 0, 1, 1); xm <-c-1.2554919, -0.6029080, -0.8017795, -0.1389861, 1.4679619, -0.8527954, 1.1096942 ,-1.0095326, 1.5500604 , 0.5622234); glm(ym~xm,family=binomial(link = "logit"))
Asked
Active
Viewed 73 times
0
-
1Well, you can start by *providing your data and your code* so we can see what's going on. – jlhoward Sep 21 '15 at 16:54
-
2Also see [here](http://stats.stackexchange.com/q/11109/5055). – joran Sep 21 '15 at 16:55