Non-parametric segmentation of non-stationary time series

TL;DR

This paper introduces a non-parametric segmentation algorithm for real-time detection of stationary intervals in non-stationary time series, improving feature uncovering beyond traditional moment-based methods.

Contribution

The authors present a novel non-parametric segmentation method capable of identifying stationary segments in non-stationary time series, applicable in real-time analysis.

Findings

Effective segmentation of diverse non-stationary time series

Uncovers features missed by moment-based methods

Applicable to real-world complex systems

Abstract

The non-stationary evolution of observable quantities in complex systems can frequently be described as a juxtaposition of quasi-stationary spells. Given that standard theoretical and data analysis approaches usually rely on the assumption of stationarity, it is important to detect in real time series intervals holding that property. With that aim, we introduce a segmentation algorithm based on a fully non-parametric approach. We illustrate its applicability through the analysis of real time series presenting diverse degrees of non-stationarity, thus showing that this segmentation procedure generalizes and allows to uncover features unresolved by previous proposals based on the discrepancy of low order statistical moments only.