TY - GEN
T1 - Consistent HMM parameter estimation using Kerridge inaccuracy rates
AU - Molloy, Timothy L.
AU - Ford, Jason J.
PY - 2013
Y1 - 2013
N2 - In this paper, we propose a novel online hidden Markov model (HMM) parameter estimator based on Kerridge inaccuracy rate (KIR) concepts. Under mild identifiability conditions, we prove that our online KIR-based estimator is strongly consistent. In simulation studies, we illustrate the convergence behaviour of our proposed online KIR-based estimator and provide a counter-example illustrating the local convergence properties of the well known recursive maximum likelihood estimator (arguably the best existing solution).
AB - In this paper, we propose a novel online hidden Markov model (HMM) parameter estimator based on Kerridge inaccuracy rate (KIR) concepts. Under mild identifiability conditions, we prove that our online KIR-based estimator is strongly consistent. In simulation studies, we illustrate the convergence behaviour of our proposed online KIR-based estimator and provide a counter-example illustrating the local convergence properties of the well known recursive maximum likelihood estimator (arguably the best existing solution).
UR - http://www.scopus.com/inward/record.url?scp=84893311655&partnerID=8YFLogxK
U2 - 10.1109/AUCC.2013.6697250
DO - 10.1109/AUCC.2013.6697250
M3 - Conference contribution
AN - SCOPUS:84893311655
SN - 9781479924981
T3 - 2013 3rd Australian Control Conference, AUCC 2013
SP - 73
EP - 78
BT - 2013 3rd Australian Control Conference, AUCC 2013
T2 - 2013 3rd Australian Control Conference, AUCC 2013
Y2 - 4 November 2013 through 5 November 2013
ER -