Maximum likelihood estimation in hidden Markov models with inhomogeneous noise

¹ Institute for Mathematical Stochastics, Georg-August-University of Göttingen, Goldschmidtstraße 7, 37077 Göttingen, Germany.
² Max Planck Institute for Biophysical Chemistry, Am Faßberg 11, 37077 Göttingen, Germany.
³ Felix-Bernstein-Institute for Mathematical Statistics in the Biosciences, Goldschmidtstraße 7, 37077 Göttingen, Germany.

^* Corresponding author: daniel.rudolf@uni-goettingen.de

Received: 11 April 2018
Accepted: 19 September 2018

Abstract

We consider parameter estimation in finite hidden state space Markov models with time-dependent inhomogeneous noise, where the inhomogeneity vanishes sufficiently fast. Based on the concept of asymptotic mean stationary processes we prove that the maximum likelihood and a quasi-maximum likelihood estimator (QMLE) are strongly consistent. The computation of the QMLE ignores the inhomogeneity, hence, is much simpler and robust. The theory is motivated by an example from biophysics and applied to a Poisson- and linear Gaussian model.