Convergence analysis of partitioned adaptive estimators under continuous parameter uncertainty

The asymptotic behavior of a Bayes Optimal adaptive estimation scheme (also called the partitioned adaptive estimation algorithm) for a sampled stochastic process with unknown parameters is investigated. The unknown parameter vector is assumed to be continuous and to belong to a compact subset of a metric space. The results are then used to analyze a scalar linear Gauss-Markov dynamical system. The multivariate linear Gauss-Markov model is also discussed.