The Centre for the Analysis of Time Series closed on 31 December 2020.
Members of CATS have now joined the LSE Data Science Institute, Department of Statistics and Grantham Institute.
These pages are kept for historical purposes.
The Centre for the Analysis of Time Series (CATS) was established in 2000 and is based within the Department of Statistics at LSE. It takes a broad definition of 'time series' and has developed a focus on predictability in theory, in practice, and in decision support.
CATS is led by Director Professor Leonard Smith with Chair Professor Henry Wynn and Co-Directors Professor Pauline Barrieu, Professor Roman Frigg and Dr David Stainforth. It hosts a variety of internationally renowned Visiting Professors, both Academic and in Practice. The School has a long and distinguished history in time series analysis and has invested heavily in developing a world-class centre of excellence in this area as part of its strategic plan.
Broadly, the Centre for the Analysis of Time Series aims to:
- Address the question of data analysis using both physical insight and the most relevant statistical methods.
- Exploit non-linear analysis in situations of economic and physical interest, including the role of forecasts both in weather-like and climate-like situations.
- Promote awareness of limitations of non-linear analysis and the danger of blindly transferring well-known physics to simulation modelling.
- Focus on end-to-end forecasting, taking account of current uncertainty about the state of the system, model inadequacy and finite computational power.
- Enjoy the beauty of purely mathematical challenges.
Suggestions for new areas of interest are always welcome. We are interested in the development of tools to interpret, value and apply probabilistic forecasts where they can be made, alternatives where they are better not made, and the clear communication of insight in each case.