Premium
Estimating the order of a hidden markov model
Author(s) -
Mackay Rachel J.
Publication year - 2002
Publication title -
canadian journal of statistics
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.804
H-Index - 51
eISSN - 1708-945X
pISSN - 0319-5724
DOI - 10.2307/3316097
Subject(s) - identifiability , hidden markov model , hidden semi markov model , markov model , context (archaeology) , variable order markov model , computer science , markov chain , forward algorithm , set (abstract data type) , machine learning , mathematics , algorithm , econometrics , artificial intelligence , geography , programming language , archaeology
While the estimation of the parameters of a hidden Markov model has been studied extensively, the consistent estimation of the number of hidden states is still an unsolved problem. The AIC and BIC methods are used most commonly, but their use in this context has not been justified theoretically. The author shows that for many common models, the penalized minimum‐distance method yields a consistent estimate of the number of hidden states in a stationary hidden Markov model. In addition to addressing the identifiability issues, she applies her method to a multiple sclerosis data set and assesses its performance via simulation.