Compression-based methods for nonparametric prediction and estimation of some characteristics of time series

Authors:
Boris Ryabko
Affiliations:
Institute of Computational Technology, Siberian Branch of the Russian Academy of Science, Siberian State University of Telecommunications and Informatics, Novosibirsk, Russia
Venue:
IEEE Transactions on Information Theory
Year:
2009

Citing 20
Cited 3

The complexity and effectiveness of prediction algorithms

Journal of Complexity
Probability (2nd ed.)

Probability (2nd ed.)
Average Case Analysis of Algorithms on Sequences

Average Case Analysis of Algorithms on Sequences
Universal Compression and Retrieval

Universal Compression and Retrieval
Information Theory and Reliable Communication

Information Theory and Reliable Communication
Information Theory: Coding Theorems for Discrete Memoryless Systems

Information Theory: Coding Theorems for Discrete Memoryless Systems
Experimental investigation of forecasting methods based on data compression algorithms

Problems of Information Transmission
Algorithmic Clustering of Music Based on String Compression

Computer Music Journal
The Minimum Description Length Principle (Adaptive Computation and Machine Learning)

The Minimum Description Length Principle (Adaptive Computation and Machine Learning)
Suboptimal behavior of Bayes and MDL in classification under misspecification

Machine Learning
Information and Complexity in Statistical Modeling

Information and Complexity in Statistical Modeling
Weakly convergent nonparametric forecasting of stationary time series

IEEE Transactions on Information Theory
Memory-universal prediction of stationary random processes

IEEE Transactions on Information Theory
Limits to consistent on-line forecasting for ergodic time series

IEEE Transactions on Information Theory
The interactions between ergodic theory and information theory

IEEE Transactions on Information Theory
Universal schemes for learning the best nonlinear predictor given the infinite past and side information

IEEE Transactions on Information Theory
Grammar-based codes: a new class of universal lossless source codes

IEEE Transactions on Information Theory
A probabilistic approach to some asymptotics in noiseless communication

IEEE Transactions on Information Theory
On optimal sequential prediction for general processes

IEEE Transactions on Information Theory
Clustering by compression

IEEE Transactions on Information Theory

Nonparametric statistical inference for ergodic processes

IEEE Transactions on Information Theory
Cloudy: heterogeneous middleware for in time queries processing

Proceedings of the 17th International Database Engineering & Applications Symposium
A binary-classification-based metric between time-series distributions and its use in statistical and learning problems

The Journal of Machine Learning Research

Quantified Score

Hi-index	754.90

Visualization

Abstract

We address the problem of online prediction for time series. We show that any universal code (or a universal data compressor) can be used as a basis for constructing asymptotically optimal methods for this problem for a certain class of stationary and ergodic processes.