BLAMJLWY@UIAMVS.BITNET (Myoung-jin Lee) (02/28/90)
Hi! Actually, I'm not so good at statistics. However, I have a little experience like that kind of problem. I think it will be very helpful to consider casual analysis involving loglinear model. For the usage of that model, there is a good introductory book. S.E. Fienberg, The Analysis of Cross-Categorical Data 1980. MIT. I hope it will be helpful.