Cyclicality, Periodicity and the Topology of Time Series

Abstract

Periodic and semi periodic patterns are very common in nature. In this paper we introduce a topological toolbox aiming in detecting and quantifying periodicity. The presented technique is of a general nature and may be employed wherever there is suspected cyclic behaviour in a time series with no trend. The approach is tested on a number of real-world examples enabling us to consistently demonstrate an ability to recognise periodic behaviour where conventional techniques fail to do so. Quicker to react to changes in time series behaviour, and with a high robustness to noise, the toolbox offers a powerful way to deeper understanding of time series dynamics.