Coarse-grained embeddings of time series: Random walks, Gaussian random processes, and deterministic chaos (Q2366507)

Whenever a time series occurs, it is important to know whether it is purely random or is generated by a deterministic process. In this paper the authors study time series by studying the statistics indicating the degree to which trajectories passing through a small region of an embedding space are parallel. By comparing results for known random and deterministic processes, the authors demonstrate a distinct underlying difference allowing one to establish a test for classifying time processes into purely random and deterministic categories. The work is empirical rather than deductive.