English

Forward estimation for ergodic time series

Probability 2015-05-13 v1 Information Theory math.IT

Abstract

The forward estimation problem for stationary and ergodic time series {Xn}n=0\{X_n\}_{n=0}^{\infty} taking values from a finite alphabet X{\cal X} is to estimate the probability that Xn+1=xX_{n+1}=x based on the observations XiX_i, 0in0\le i\le n without prior knowledge of the distribution of the process {Xn}\{X_n\}. We present a simple procedure gng_n which is evaluated on the data segment (X0,...,Xn)(X_0,...,X_n) and for which, error(n)=gn(x)P(Xn+1=xX0,...,Xn)0{\rm error}(n) = |g_{n}(x)-P(X_{n+1}=x |X_0,...,X_n)|\to 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.

Cite

@article{arxiv.0711.3856,
  title  = {Forward estimation for ergodic time series},
  author = {Gusztav Morvai and Benjamin Weiss},
  journal= {arXiv preprint arXiv:0711.3856},
  year   = {2015}
}
R2 v1 2026-06-21T09:46:55.755Z