Detecting chaos in pseudoperiodic time series without embedding

Abstract

A different method is proposed to detect deterministic structure from a pseudoperiodic time series. By using the correlation coefficient as a measure of the distance between cycles, we are exempt from phase-space reconstruction and further construct a hierarchy of pseudocycle series that, in turn, preserve less determinism than the original time series. Appropriate statistics are then devised to reveal the temporal and spatial correlation encoded in this hierarchy of the pseudocycle series, which allows for a reliable detection of determinism and chaos in the original time series. We demonstrate that this method can reliably identify chaos in the presence of noise of different sources for both artificial data and experimental time series.