Comments on model validation as set membership identification

Abstract

We review four basic model validation techniques, one that relates to the “unknown-but-bounded” disturbance assumption, one that has been recently suggested in the “Identification-for-robust-control” context and two more classical statistical tests. By defining the set of models that would pass the chosen model validation test, we may interpret each of these as a set membership identification method. The consequences of such a viewpoint are discussed, and we focus on the important, but perhaps controversial concept of “independence” to make further selections of models within the thus defined sets.