RECURSIVE MODEL ORDER DETERMINATION.

Abstract

Two algorithms are proposed to compute recursively the model order testing criteria and parameters simultaneously for K nested models which have similar structures but with increasing complexity where K is the highest model order being considered. The algorithms exploit certain properties of the U**T DU factorization of the augmented covariance matrix and the UDU**T factorisation of the augmented information matrix corresponding to the parameters. It is shown that the parameters corresponding to the different model orders are contained in the U factor while several of the well established model order testing criteria for different models can readily be computed from the D factor.