Forward estimation for ergodic time series
Gusztáv Morvai, Benjamin Weiss, Ann Inst, H. Poincaré, Probab Statist · 2005
The forward estimation problem for stationary and ergodic time series {Xn} ∞ n=0 taking values from a finite alphabet X is to estimate the probability that Xn+1 = x based on the observations Xi, 0 ≤ i ≤ n without prior knowledge of the distribution of the process {Xn}. We present a simple procedure gn which is evaluated on the data segment (X0,...,Xn) and for which, error(n) = |gn(x) − P(Xn+1 = x|X0,...,Xn) | → 0 almost surely for a subclass of all stationary and ergodic time series, while for the full class the Cesaro average of the error tends to zero almost surely and moreover, the error tends to zero in probability. Le problème d’estimation future d’une série de temps ergodique et stationnaire {Xn} ∞ n=0, qui prend ses valeures dans un alphabet fini X, est d’estimer la probabilité que Xn+1 = x, connaissant les Xi pour 0 ≤ i ≤ n mais sans connaissance préalable de la distribution du