Hi all, I'm currently comparing several multiple linear regression models and want to see if there are statistically significant differences between the models, say whether one model is significantly better than another. This might be a very simple question but I just cannot find the most appropriate method. The dependent variables and samples are exactly the same for all the models, while independent variables are overlapping. It seems that likelihood ratio test can only be applied to nested models, and Vuong test/Clarke test can only be used to strict non-nested models. Does anyone have any idea about how to compare the overlapping models using SPSS or other softwares? Thanks a lot! Best regards, Shu PhD student UCL Energy Institute 14 Upper Woburn Place, WC1H 0NN London, UK You may leave the list at any time by sending the command SIGNOFF allstat to [log in to unmask], leaving the subject line blank.