On universal algorithms for classifying and predicting stationary processes

This is a survey of results on universal algorithms for classification and prediction stationary processes. The problems include discovering the order k-step Markov chain, determining memory words in finitarily Markovian processes estimating entropy an unknown process. cover both discrete real valued variety situations. Both forward backward are discussed with emphasis being pointwise results. just teaser. purpose merely to call attention prediction. We will refer interested reader sources. Throughout paper we give illuminating examples.