VARIABLE SELECTION FOR THE GROWTH CURVE MODEL

In this paper we consider the problem of selecting the covariables wit hin individuals in the growth curve model. We propose two modification s of AIC and MIC (Cp-static), which have improvements on the bias prop erties. Asymptotic distributions of variable slection criteria are der ived under a general situation where a polynomial growth curve of degr ee j(0) is approximately suitable. A simulation study is also given to gain some understanding on the small sample properties of these varia ble selection criteria. (C) 1997 Academic Press.