Cyclical ripple effects in currency markets: meso-scale pattern dynamics and temporal network analysis
Document Type
Research-Article
Journal Name
Nonlinear Dynamics
Keywords
Cyclical ripple effects, Meso-scale pattern analysis, Regional currency market, Shape-based similarity, Temporal network
Abstract
Characterizing how structured patterns emerge and associate across interconnected system components remains a fundamental challenge, particularly when dynamics are encoded in discrete, meso-scale units rather than continuous signals. This paper introduces the cyclical ripple effect, defined as the phenomenon whereby a volatility cycle in one system component is followed by a morphologically similar cycle in another with a measurable time lag. To systematically identify, quantify, and map these cyclical associations, we develop the multivariate temporal cyclical ripple effect model. It is a three-stage analytical framework that extracts cycles using an empirical mode decomposition-enhanced Bry-Boschan algorithm, quantifies ripple effects through a sliding-window shape-based distance search with temporal discounting, and constructs temporal networks to map evolving association topologies. Applying the framework to 15 Regional Comprehensive Economic Partnership (RCEP) currencies from 2013 to 2023, we find that aggregate ripple effect intensity surged during acute external shocks and dropped during prolonged uncertainty and adjustment. The temporal network reveals a persistent positional hierarchy, with free-float currencies tend to act as consistent leaders and managed-pegged currencies as followers. Notably, the Chinese Yuan transitions from dominant leader in 2016 to central intermediary during 2018–2019, coinciding with post-reform adjustment and trade tensions. The framework contributes a replicable methodology for analyzing meso-scale pattern interdependence in multivariate time series. © The Author(s), under exclusive licence to Springer Nature B.V. 2026.