Hidden Markov Chain Filtering for Generalised Bessel Processes
Finite-dimensional recursive filters are obtained for generalised Bessel processes with a drift parameter that follows a hidden Markov chain. In particular, filters are constructed for the states, the jumps and the occupation times of the states of the Markov chain. These lead to estimators for the transition rates and the levels of the hidden states of the chain. Finally a minimum variance filter is described that minimises fluctuations of the filters.