Title :
Estimation of Markov-modulated time-series via EM algorithm
Author :
Dey, Subhrakanti ; Krishnamurthy, Vikram ; Salmon-Legagneur, Thierry
Author_Institution :
Dept. of Syst. Eng., Australian Nat. Univ., Canberra, ACT, Australia
Abstract :
We consider the estimation of various Markov-modulated time series. We obtain maximum likelihood estimates of the time-series parameters including the Markov chain transition probabilities and the time-series coefficients using the expectation maximization (EM) algorithm. In addition, the recursive EM algorithm is used to obtain on-line parameter estimates. Simulation studies show that both algorithms yield satisfactory results.<>
Keywords :
Markov processes; maximum likelihood estimation; probability; signal processing; time series; EM algorithm; Markov chain transition probabilities; Markov-modulated time-series; expectation maximization algorithm; maximum likelihood estimates; on-line parameter estimates; recursive EM algorithm; simulation studies; time-series coefficients; time-series parameters; Adaptive systems; Australia; Econometrics; Image segmentation; Maximum likelihood estimation; Parameter estimation; Polynomials; Recursive estimation; Robustness; Signal processing algorithms;
Journal_Title :
Signal Processing Letters, IEEE
