Sequential Prediction of Unbounded Stationary Time Series
Authors:
- 30 April 2007
- journal article
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Information Theory
- Vol. 53 (5) , 1866-1872
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 processesKeywords
This publication has 22 references indexed in Scilit:
- Universal prediction of random binary sequences in a noisy environmentThe Annals of Applied Probability, 2004
- On optimal sequential prediction for general processesIEEE Transactions on Information Theory, 2003
- A simple randomized algorithm for sequential prediction of ergodic time seriesIEEE Transactions on Information Theory, 1999
- Universal predictionIEEE Transactions on Information Theory, 1998
- Memory-universal prediction of stationary random processesIEEE Transactions on Information Theory, 1998
- Weakly convergent nonparametric forecasting of stationary time seriesIEEE Transactions on Information Theory, 1997
- Nonparametric inference for ergodic, stationary time seriesThe Annals of Statistics, 1996
- The strong law of large numbers for sequential decisions under uncertaintyIEEE Transactions on Information Theory, 1994
- Universal prediction of individual sequencesIEEE Transactions on Information Theory, 1992
- Universal Schemes for Prediction, Gambling and Portfolio SelectionThe Annals of Probability, 1992