Abstract

We revisit the decay of missing ordinal patterns in long-term correlated time series. More precisely, a stretched exponential model is proposed to describe more appropriately how the number of missing ordinal patterns decreases as a function of the time series length in the case of correlated stochastic data. Numerical analysis and experimental applications confirm that this generalized model allows a more accurate characterization of the underlying dynamics and a reliable quantification of the long-range temporal correlations. (C) 2019 Elsevier B.V. All rights reserved.