Prediction for discrete time series
Abstract
Description
Let $\{X_n\}$ be a stationary and ergodic time series taking values from a finite or countably infinite set ${\cal X}$. Assume that the distribution of the process is otherwise unknown. We propose a sequence of stopping times $λ_n$ along which we will be able to estimate the conditional probability $P(X_{λ_n+1}=x|X_0,...,X_{λ_n})$ from data segment $(X_0,...,X_{λ_n})$ in a pointwise consistent way for a restricted class of stationary and ergodic finite or countably infinite alphabet time series which includes among others all stationary and ergodic finitarily Markovian processes. If the stationary and ergodic process turns out to be finitarily Markovian (among others, all stationary and ergodic Markov chains are included in this class) then $ \lim_{n\to \infty} {n\over λ_n}>0$ almost surely. If the stationary and ergodic process turns out to possess finite entropy rate then $λ_n$ is upperbounded by a polynomial, eventually almost surely.