Identification of Continuous-Discrete Hidden Markov Models with Multiplicative Observation Noise

The paper aims to identify hidden Markov model parameters. unobservable state represents a finite-state jump process. observations contain Wiener noise with state-dependent intensity. identified parameters include the transition intensity matrix of system state, conditional drift and diffusion coefficients in observations. We propose an iterative identification algorithm based on fixed-interval smoothing state. Using calculated estimates, we restore all required contains detailed description numerical schemes estimation parameter identification. comprehensive study confirms high precision proposed estimates.