Analysis on the performance and properties of interacting multiple models algorithm (Q2785168)

The present paper is devoted to the study of the interacting multiple models algorithm. The authors present a nonsimulation method to analyse the Markov transition probability matrix with \(m\) parameters of an interacting multiple model estimator.NEWLINENEWLINENEWLINESeveral conclusions independent of the scenarios and model-conditional filters are given. Part of the analysis shown herein is also suitable for a first-order generalized pseudo-Bayesian algorithm.NEWLINENEWLINENEWLINEThis paper consists of five sections: 1. Introduction; 2. Interacting multiple models algorithm; 3. Effect Markov parameters on model probability; 4. Effect analysis of Markov parameters on model-conditional estimation; and 5. Conclusion.