Compression-Based Methods for Nonparametric Prediction and Estimation of Some Characteristics of Time Series
Boris Ya. Ryabko · IEEE Transactions on Information Theory · 2009
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.