K.U.Leuven - Faculty of Economics and Applied Economics
DTEW - KBI_0715 pages:1-21
Application of classical model selection methods such as Akaike’s information criterion AIC becomes problematic when observations are missing. In this paper we propose some variations on the AIC, which are applicable to missing covariate problems. The method is directly based on the EM algorithm and is readily available for EM-based estimation methods, without much additional computational efforts. The missing data-AIC criteria are formally derived and shown to work in a simulation study and by application to data on diabetic retinopathy.