I do have a database of 6 features. I would like to compare LDA, and PC-LDA. I can evaluate the number of components in PCA but failed to find in LDA. Any kind of help is appreciable.
Hi, what do yo mean by "evaluating the number of components"? In LDA, the number of discriminant functions is the number of groups - 1. So, if you have two groups, only one discriminant function allows separation between both groups.
Before you continue with your analyzes tell me what questions are you trying to answer with your data? You say you have 6 variables. Right? What hypotheses would you like to answer?