Recursive estimation of multivariate hidden Markov model parameters
| Autor: | Leonidas Sakalauskas, Jūratė Vaičiulytė |
|---|---|
| Rok vydání: | 2019 |
| Předmět: |
Statistics and Probability
Estimation Multivariate statistics Computer science Estimation theory 05 social sciences Multivariate normal distribution 01 natural sciences Online analysis 010104 statistics & probability Computational Mathematics hidden Markov models Likelihood method recursive EM algorithm clustering 0502 economics and business State transition probability 0101 mathematics Statistics Probability and Uncertainty Hidden Markov model Cluster analysis Algorithm 050205 econometrics |
| Zdroj: | Computational statistics, Heidelberg : Springer, 2019, vol. 34, p. 1337-1353 |
| ISSN: | 1613-9658 0943-4062 |
| DOI: | 10.1007/s00180-019-00877-z |
| Popis: | This article addresses a recursive parameter estimation algorithm for a hidden Markov model (HMM). The work focuses on an HMM with multiple states that are assumed to follow from a multivariate Gaussian distribution. The novelty of this study lies in a state transition probability calculation technique that simplifies the application of the backward stage of the forward–backward algorithm. For sequential observation analysis, the complexity of the created recursive algorithm for learning the HMM parameters is merely linear. Meanwhile, the classical Baum–Welch algorithm has second order complexity; therefore, it cannot be applied in online analysis situations. The properties of the proposed recursive expectation–maximization (EM) algorithm were explored by a computer simulation solving test examples and demonstrate that this algorithm can be efficiently applied to solve online tasks related to HMM parameter estimation. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|
Full text from SpringerLink