Paper #507

Title:

Strategies for sequential prediction of stationary time series

Authors:

László Györfi and Gábor Lugosi

Date:

September 2000

Abstract:

We present simple procedures for the prediction of a real valued sequence. The algorithms are based on a combination of several simple predictors. We show that if the sequence is a realization of a bounded stationary and ergodic random process then the average of squared errors converges, almost surely, to that of the optimum, given by the Bayes predictor. We offer an analog result for the prediction of stationary gaussian processes.

Keywords:

Sequential prediction, ergodic process, individual sequence, gaussian process

JEL codes:

C13, C14

Area of Research:

Statistics, Econometrics and Quantitative Methods

Published in:

In Moshe Dror, Pierre L'Ecuyer, Ferenc Szidarovszky (editors), Kluwer Academic Publishers, 2001
With the title:
Modeling Uncertainty:An examination of its theory, methods, and applications (book)

Download the paper in PDF format