Title :
Particle swarm optimization particle filtering for dual estimation
Author :
Yang, Xiaojun ; Lv, Jin ; Feng, Xinle
Author_Institution :
Sch. of Inf. Eng., Chang´´an Univ., Xi´´an, China
Abstract :
A new method for the dual estimation in state-space dynamic model was described, with a focus on sequential Bayesian learning about time-varying state and static parameters simultaneously. The new algorithm combines auxiliary particle filtering (APF) with particle swarm optimization (PSO) to achieve computational efficiency and stability. By properly choosing the fitness function of PSO, the algorithm produces the recursive maximum-likelihood estimation of the parameters. It is shown that the PSO can be integrated with APF in the simulation-based sequential frame for dual estimation. The algorithm is tested on Markov switching stochastic volatility model with promising results.
Keywords :
Bayes methods; learning (artificial intelligence); maximum likelihood estimation; particle filtering (numerical methods); particle swarm optimisation; state-space methods; stochastic systems; Markov switching stochastic volatility model; computational efficiency; fitness function; particle swarm optimization particle filtering; recursive maximum likelihood estimation; sequential Bayesian learning; simulation-based sequential frame; state-space dynamic model dual estimation; time-varying state; Algorithm design and analysis; Estimation; Filtering; Heuristic algorithms; Markov processes; Monte Carlo methods; Particle swarm optimization;
Conference_Titel :
Intelligent Control and Information Processing (ICICIP), 2010 International Conference on
Conference_Location :
Dalian
Print_ISBN :
978-1-4244-7047-1
DOI :
10.1109/ICICIP.2010.5564288