Supplement to: Package AdvEMDpy: Algorithmic Variations of Empirical Mode Decomposition in Python

Abstract

This work serves as a formal supplement to ‘Package AdvEMDpy: Algorithmic Variations of Empirical Mode Decomposition in Python’ with additional synthetic and real-world examples. AdvEMDpy will be shown to be more accurate than its Python competitors in resolving the underlying driving function of the Duffing Equation, before it is used to isolate different frequency structures present in Carbon ETF data. An annual fluctuation will be extracted and possibly causally linked to the seasonal trend of the Carbon Dioxide concentration in the atmosphere. These examples are by no means exhaustive and merely serve as demonstrations of AdvEMDpy’s usage and superiority.

Keywords: Empirical Mode Decomposition (EMD), Statistical EMD (SEMD), Enhanced EMD (EEMD), Ensemble EMD, Hilbert transform, time series analysis, filtering, graduation, Winsorization, downsampling, splines, knot optimisation, Python, R, MATLAB

JEL Classification: C02, C14, C22, C32, C61, C63, C65, C88