Time series (measurements of some quantity over time) are studied in many different disciplines including medicine (ECGs, EEGs), climatology (rainfall, temperature), geology (seismic recordings, ice core data), physics (astronomical and sunspot recordings), engineering (control), economics (share prices, volume data), mathematics and statistics (dynamical systems, nonlinear dynamics and chaos), and so on. However, the methods and models used to uncover structure in these particular signals tends to be highly discipline-specific. Could it be that they are all reinventing the wheel, or is there something to be learnt by trying to compare and unify the different approaches to time-series analysis?
This seminar series aimed to promote interdisciplinary collaboration on time series, and featured speakers explaining a particular problem in their area of expertise to an interdisciplinary audience. Lessons learnt from other disciplinary perspectives were shared, leading to exciting new collaborations to direct new approaches for the field. More specifically, topics discussed included:
- time-series analysis in econometrics
- diagnosis using analysis of medical time-series
- high throughput methods for audio analysis
- geophysical models of climate variability
- astronomy and lasers: time-series from physics laboratories.
Ben Fulcher, Department of Physics and Balliol College, University of Oxford
Nick Jones, Department of Physics, University of Oxford
Andrew Whitby, Department of Economics and Balliol College, University of Oxford
Contact details for enquiries
Please email the lead investigator, Ben Fulcher, for any queries regarding this project.