Weakly Convergent Nonparametric Forecasting of Stationary Time Series

Abstract

The conditional distribution of the next outcome given the infinite past of a stationary process can be inferred from finite but growing segments of the past. Several schemes are known for constructing pointwise consistent estimates, but they all demand prohibitive amounts of input data. In this paper we consider real-valued time series and construct conditional distribution estimates that make much more efficient use of the input data. The estimates are consistent in a weak sense, and the question whether they are pointwise consistent is still open. For finite-alphabet processes one may rely on a universal data compression scheme like the Lempel-Ziv algorithm to construct conditional probability mass function estimates that are consistent in expected information divergence. Consistency in this strong sense cannot be attained in a universal sense for all stationary processes with values in an infinite alphabet, but weak consistency can. Some applications of the estimates to on-line forecasting, regression and classification are discussed.

Cited by in corpus (17)

• Universal Estimation of Directed Information
• Limits to consistent on-line forecasting for ergodic time series
• Asymptotic nonparametric statistical analysis of stationary time series
• Pattern Recognition for Conditionally Independent Data
• Inferring the conditional mean
• Nonparametric sequential prediction for stationary processes
• On universal estimates for binary renewal processes
• Forecasting for stationary binary time series
• Compression-based methods for nonparametric density estimation, on-line prediction, regression and classification for time series
• Prediction of Large Alphabet Processes and Its Application to Adaptive Source Coding
• On Sequential Estimation and Prediction for Discrete Time Series
• Strongly consistent nonparametric forecasting and regression for stationary ergodic sequences
• Intermittent estimation of stationary time series
• Regression estimation from an individual stable sequence
• Universal Codes as a Basis for Time Series Testing
• Universality of Bayesian mixture predictors
• Universal Codes as a Basis for Nonparametric Testing of Serial Independence for Time Series