Skip to main content
eScholarship
Open Access Publications from the University of California

Monte Carlo approaches to hidden Markov model state estimation

  • Author(s): Sollberger, Derek
  • Advisor(s): Bhat, Harish
  • et al.
Abstract

In this paper, we develop a Monte Carlo approach for hidden Markov model (HMM) order estimation-finding the underlying number of states in a hidden Markov model. We compare predictions and true observations using classification rates, correlations, and ROC curves as statistical estimators. Tests are run on both artificail data in a controlled experiment and on real-world data sets--Abstract.

Main Content
Current View