Limitations on intermittent forecasting

DOI10.1016/J.SPL.2004.12.016MaRDI QIDQ2483866zbMATH OpenOpenAlexFDO

Authors Gusztáv Morvai, Benjamin Weiss

Publication date 1 August 2005

Published in Statistics \& Probability Letters (Search for Journal in Brave)

Full work available at URL https://arxiv.org/abs/0710.3773

zbMATH Keywords

Finite-order Markov chainsPrediction theoryStationary and ergodic processes

Mathematics Subject Classification ID

Stationary stochastic processes (60G10) Prediction theory (aspects of stochastic processes) (60G25) Stopping times; optimal stopping problems; gambling theory (60G40) Markov chains (discrete-time Markov processes on discrete state spaces) (60J10) Nonparametric estimation (62G05) Optimal stopping in statistics (62L15) Time series, auto-correlation, regression, etc. in statistics (GARCH) (62M10) Inference from stochastic processes and prediction (62M20)

Abstract: Bailey showed that the general pointwise forecasting for stationary and ergodic time series has a negative solution. However, it is known that for Markov chains the problem can be solved. Morvai showed that there is a stopping time sequence $lambd{a}_n$ such that $P(X_{lambd{a}_n+1}=1|X_0,\mathrm{...},X_{lambd{a}_n})$ can be estimated from samples $(X_0,\mathrm{...},X_{lambd{a}_n})$ such that the difference between the conditional probability and the estimate vanishes along these stoppping times for all stationary and ergodic binary time series. We will show it is not possible to estimate the above conditional probability along a stopping time sequence for all stationary and ergodic binary time series in a pointwise sense such that if the time series turns out to be a Markov chain, the predictor will predict eventually for all $n$.

Recommendations

• Intermittent estimation of stationary time series
• Forecasting for stationary binary time series
• A note on prediction for discrete time series
• Prediction for discrete time series
• Limits to consistent on-line forecasting for ergodic time series

Cites work

• Forecasting for stationary binary time series
• scientific article; zbMATH DE number 5280104 (Why is no real title available?)
• scientific article; zbMATH DE number 1399413 (Why is no real title available?)
• Large-scale typicality of Markov sample paths and consistency of MDL order estimators
• Limits to consistent on-line forecasting for ergodic time series
• Prediction of random sequences and universal coding
• Random Markov processes and uniform martingales
• Strongly mixing g-measures
• The consistency of the BIC Markov order estimator.

Cited in

(9)

• Forecasting for stationary binary time series
• On universal algorithms for classifying and predicting stationary processes
• Intermittent estimation of stationary time series
• ON SEQUENTIAL ESTIMATION AND PREDICTION FOR DISCRETE TIME SERIES
• Limits to consistent on-line forecasting for ergodic time series
• Consistency, integrability and asymptotic normality for some intermittent estimators
• Nonparametric sequential prediction for stationary processes
• Countable alphabet stationary processes with at least one memory word and intermittent estimation with universal rates
• On universal estimates for binary renewal processes

This page was built for publication: Limitations on intermittent forecasting

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q2483866)