Navigation

External Pages

Time Series Analysis using Empirical Mode Decomposition and Hilbert-Huang Transform

Brent Davis and Nathan Taylor, Cal Poly, Mathematics Dept.


Abstract:

Empirical Mode Decomposition is a time series technique created in 1999. It is a decomposition method which separates a time series into a finite set of intrinsic mode functions (IMF). IMFs could have physical meaning in applications. The talk introduces and explains Empirical Mode Decomposition and explains the mathematical artifacts of the technique and its advantage over traditional Power Spectrum Analysis. It is then applied to paleoclimatological data to draw correlations with the Milankovitch cycles of the earths movement.