208
v1v2 (latest)

Uniform hypothesis testing for ergodic time series distributions

IEEE Region International Conference on Computational Technologies in Electrical and Electronics Engineering (ICCTEEE), 2010
Abstract

Given a discrete-valued sample X1,...,XnX_1,...,X_n we wish to decide whether it was generated by a distribution belonging to a family H0H_0, or it was generated by a distribution belonging to a family H1H_1. In this work we assume that all distributions are stationary ergodic, and do not make any further assumptions (e.g. no independence or mixing rate assumptions). We would like to have a test whose probability of error (both Type I and Type II) is uniformly bounded. More precisely, we require that for each ϵ\epsilon there exist a sample size nn such that probability of error is upper-bounded by ϵ\epsilon for samples longer than nn. We find some necessary and some sufficient conditions on H0H_0 and H1H_1 under which a consistent test (with this notion of consistency) exists. These conditions are topological, with respect to the topology of distributional distance.

View on arXiv
Comments on this paper