- Organisation & People
- Research & Publications
- PhD Programme
- PhD Roadmap, TSP & Diploma
- PhD Graduations
- News & Events
Researchers with data collected regularly over a time period. Must already have a working knowledge of Matlab or R (e.g. read/write data, work with matrices).
The course provides a practical introduction to time series analyses and covers three main topics:
1) Data preparation and visualization, dealing with data gaps and methods for interpolation.
2) Finding trends and periodicities using empirical mode decomposition; methods for noise reduction and auto correlation analysis.
3) Methods for comparing two or more time series with respect to similarities, asynchronism, distance measures.
Understand what are time series, stationarity, Fourier transform, distance matrices. Perform data interpolation, empirical mode decomposition, cross-correlation, time series clustering, and dynamic time warping.
> More information, Module 2017-15