Summary
2007 International Symposium on Nonlinear Theory and its Applications
2007
Session Number:18AM2-B
Session:
Number:18AM2-B-1
Prediction of high-dimensional multivariate information as an amplitude-event dynamical system
Naoki Yabuta, Tohru Ikeguchi,
pp.188-191
Publication Date:2007/9/16
Online ISSN:2188-5079
DOI:10.34385/proc.41.18AM2-B-1
Summary:
To predict high dimensional multivariate information, such as event sizes and timings, conventional frameworks have usually treated event sizes or event timings separately. However, in real worlds, we often able to measure the event sizes and the event timings simultaneously. Therefore, it is very natural to consider that if we can measure them, it might be better to handle them simultaneously in or- der to realize higher predictability. Following the idea, a new prediction framework using both event sizes and tim- ings has already been proposed. However, the framework was applied only to a low-dimensional chaos. Therefor, in the present paper, we applied the framework to high- dimensional chaos with a nonlinear prediction method. As a result, it is shown that the framework exhibits high perfor- mance than the conventional framework even if it is applied to high-dimensional chaos.