Catalogo Articoli (Spogli Riviste)
OPAC HELP
Titolo: Reconstructions and predictions of nonlinear dynamical systems: A hierarchical Bayesian approach
Autore: Matsumoto, T; Nakajima, Y; Saito, M; Sugi, J; Hamagishi, H;
 Indirizzi:
 Waseda Univ, Dept Elect Elect & Comp Engn, Tokyo, Japan Waseda Univ Tokyo Japan niv, Dept Elect Elect & Comp Engn, Tokyo, Japan Japan Sci & Technol Corp, Core Res Evolut Sci & Technol, Tokyo, Japan Japan Sci & Technol Corp Tokyo Japan Evolut Sci & Technol, Tokyo, Japan
 Titolo Testata:
 IEEE TRANSACTIONS ON SIGNAL PROCESSING
fascicolo: 9,
volume: 49,
anno: 2001,
pagine: 2138  2155
 SICI:
 1053587X(200109)49:9<2138:RAPOND>2.0.ZU;20
 Fonte:
 ISI
 Lingua:
 ENG
 Soggetto:
 SERIES;
 Keywords:
 hierarchical Bayesian approach; neural net; nonlinear time series prediction;
 Tipo documento:
 Article
 Natura:
 Periodico
 Settore Disciplinare:
 Engineering, Computing & Technology
 discip_EC
 Citazioni:
 36
 Recensione:
 Indirizzi per estratti:
 Indirizzo: Matsumoto, T Waseda Univ, Dept Elect Elect & Comp Engn, Tokyo, Japan Waseda Univ Tokyo Japan ct Elect & Comp Engn, Tokyo, Japan



 Citazione:
 T. Matsumoto et al., "Reconstructions and predictions of nonlinear dynamical systems: A hierarchical Bayesian approach", IEEE SIGNAL, 49(9), 2001, pp. 21382155
Abstract
An attempt is made to reconstruct model nonlinear dynamical systems from scalar time series data via a hierarchical Bayesian framework. Reconstruction is performed by fitting given training data with a parameterized family of functions without overfitting. The reconstructed model dynamical systems are compared with respect to (approximated) model marginal likelihood, which is a natural Bayesian information criterion. The best model is selected with respect to this criterion and is utilized to make predictions. The results are applied to two problems: i) Chaotic time series prediction and ii) building airconditioning load prediction. The former is a very good class of problems for checking abilities of prediction algorithms for at least two reasons. First, since no linear dynamical systems can admit chaotic behavior, an algorithm must capture nonlinearities behind the time series. Second, chaotic dynamical systems are sensitive to initial conditions. More precisely, the error grows exponentially with respect to time so that crispnessof capturing nonlinearities is also important. Experimental results appearto indicate that the proposed scheme can capture difficult nonlinearities behind chaotic time series data. The latter class of problems (air conditioning load prediction) is motivated by a great amount of demand for reducingCO2 emissions associated with electric power generation. The authors won aprediction competition using the proposed algorithm; therefore, it appearsto be reasonably sound.
ASDD Area Sistemi Dipartimentali e Documentali, Università di Bologna, Catalogo delle riviste ed altri periodici
Documento generato il 05/07/20 alle ore 04:58:34