I'm trying to run a model using Proc Logistic but I get an error message saying "no valid observations". I read that including a variable whose values are 100% missing in a proc logistic model results in this error. I went back and found that some of my variables did indeed have values that were 95 - 100% missing, so I took them out of the model but I'm still getting this error message. Can someone tell me any other reasons as to what might be causing this?? ... View more