Blind Construction of Optimal Nonlinear Recursive Predictors for Discrete Sequences

zbMATH Open1073.68516arXivcs/0406011MaRDI QIDQ4451013FDOQ4451013

Cosma Rohilla Shalizi

Publication date: 22 February 2004

Abstract: We present a new method for nonlinear prediction of discrete random sequences under minimal structural assumptions. We give a mathematical construction for optimal predictors of such processes, in the form of hidden Markov models. We then describe an algorithm, CSSR (Causal-State Splitting Reconstruction), which approximates the ideal predictor from data. We discuss the reliability of CSSR, its data requirements, and its performance in simulations. Finally, we compare our approach to existing methods using variable-length Markov models and cross-validated hidden Markov models, and show theoretically and experimentally that our method delivers results superior to the former and at least comparable to the latter.

Full work available at URL: https://arxiv.org/abs/cs/0406011

zbMATH Keywords

networksrecursive estimationrandom fieldsinformation theorysufficient statisticsnonlinear prediction

Mathematics Subject Classification ID

Network design and communication in computer systems (68M10)

Cited In (3)

• Application of computational mechanics to the analysis of seismic time-series via numerical optimisation
• Fast Non-mean-field Networks: Uniform in Time Averaging
• A location-mixture autoregressive model for online forecasting of lung tumor motion

   Recommendations

• Nonparametric prediction for random fields 👍 👎
• Title not available (Why is that?) 👍 👎
• Nonlinear prediction of a class of random processes 👍 👎
• Predicting random fields with increasing dense observations 👍 👎
• A Neural Net Model for Prediction 👍 👎

This page was built for publication: Blind Construction of Optimal Nonlinear Recursive Predictors for Discrete Sequences

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q4451013)