Combining the ApEn statistic with surrogate data analysis for the detection of nonlinear dynamics in time series

We tested the natural combination of surrogate data analysis with the ApEn regularity statistic developed by Pincus [Proc. Natl. Acad. Sci. USA 88 (1991) 2297] by applying it to some popular models of nonlinear dynamics and publicly available experimental time series. We found that this easily implemented combination provided a useful method for discriminating signals governed by nonlinear dynamics from those governed by linear dynamics and noise. An apparently novel physical interpretation of ApEn also is supplied.