04-03-2017 09:49 AM
What can i do if i have a data set which its residuals are not normally distributed even after applying all types of transformations in SAS. I understand to avoid biase i have to use the NPAR1WAY rather than the GLIMMIX procedure but since i do not have XLSTAT in my PC is there anyway i can further do the multiple comparison in SAS?. From the results i need the F- value with its numerator/denominator degrees of freedom and the P value.
Any help would be greatly appreciated!