Selecting Statistical Models with Information Statistics.

                                                   YASUHIKO WADA
                           National Institute of Animal Industry
                                             Tsukuba Norindanchi
                                              Ibaraki 305, Japan

                                              NOBUHISA KASHIWAGI
                        The Institute of Statistical Mathematics
                                         Minami Azabu, Minato-ku
                                                Tokyo 106, Japan

ABSTRACT

  The problem of model selection in animal breeding is examined
using Akaike's Information Criterion. A detailed procedure is
presented for the mixed linear model. A simulation study was
designed based on field data on Japanese Holstein cows. Criteria
related to Akaike's information criterion, extensions of the 
proposed procedure, and computational difficulties are discussed.

Key words

best linear unbiased preocedure, Akaike's information criterion, model selection
                              J. Dairy Sci. 73: 3575-3582. (1990)

Wada's Lab.