Model Selection and Criticism

Abstract

In many applications of statistics, little prior knowledge or relevant theory is available, and so model choice becomes an entirely empirical, exploratory process. Two different procedures are described in this chapter. The first is stepwise selection in which, starting from some initial model, edges are successively added or removed until some criterion is fulfilled. The second is a more global search technique proposed by Edwards and Havránek (1985, 1987) which seeks the simplest models consistent with the data. In the third section of this chapter a brief comparison of the two approaches is made.