Title :
Bayesian MCMC nonlinear time series prediction: predictive mean and error bar
Author :
Nakada, Yohei ; Matsumoto, Takashi
Author_Institution :
Dept. of Electr., Electron. & Comput. Eng., Waseda Univ., Tokyo, Japan
Abstract :
When nonlinear dynamical system is behind time series data, predictions are rather difficult. Hierarchical Bayesian scheme previously proposed by the authors has been shown to be reasonably sound. A great difficulty implementing the hierarchical Bayesian scheme lies in the computation of predictive distributions for which quadratic approximations have been used so far. This paper attempts to compute predictive mean and error bar for nonlinear time series prediction problems via MCMC (Markov Chain Monte Carlo) without quadratic approximations. The scheme is tested against time series generated by (Chaotic) Rossler system
Keywords :
Bayes methods; Markov processes; Monte Carlo methods; time series; Markov chain Monte Carlo method; error bar; hierarchical Bayesian scheme; nonlinear dynamical system; nonlinear time series prediction problems; predictive distributions; predictive mean; quadratic approximations; Acoustical engineering; Bayesian methods; Chaos; Computer errors; Markov processes; Monte Carlo methods; Neural networks; Nonlinear dynamical systems; Predictive models; Uncertainty;
Conference_Titel :
Neural Networks for Signal Processing X, 2000. Proceedings of the 2000 IEEE Signal Processing Society Workshop
Conference_Location :
Sydney, NSW
Print_ISBN :
0-7803-6278-0
DOI :
10.1109/NNSP.2000.889406