DocumentCode
818372
Title
Sequential Prediction of Unbounded Stationary Time Series
Author
Györfi, László ; Ottucsák, György
Author_Institution
Dept. of Comput. Sci. & Inf. Theor., Budapest Univ. of Technol. & Econ.
Volume
53
Issue
5
fYear
2007
fDate
5/1/2007 12:00:00 AM
Firstpage
1866
Lastpage
1872
Abstract
A simple on-line procedure is considered for the prediction of a real-valued sequence. The algorithm is based on a combination of several simple predictors. If the sequence is a realization of an unbounded stationary and ergodic random process then the average of squared errors converges, almost surely, to that of the optimum, given by the Bayes predictor. An analog result is offered for the classification of binary processes
Keywords
Bayes methods; binary sequences; pattern classification; prediction theory; random processes; time series; Bayes predictor; binary process; ergodic random process; real-valued sequential prediction; stationary time series; Automation; Convergence; Economic forecasting; High-speed networks; Informatics; Pattern recognition; Random processes; Random variables; Statistical learning; Telecommunication computing; On-line learning; pattern recognition; sequential prediction; time series; universal consistency;
fLanguage
English
Journal_Title
Information Theory, IEEE Transactions on
Publisher
ieee
ISSN
0018-9448
Type
jour
DOI
10.1109/TIT.2007.894660
Filename
4167734
Link To Document