The LSE has a long and distinguished history in time series analysis. At present, the Department's time series group consists of Professors Leonard Smith|, Piotr Fryzlewicz| and Qiwei Yao| and Lecturers Dr Konstantinos Kalogeropoulos|, Dr Clifford Lam| and Dr Matteo Barigozzi|. Our research is greatly enhanced by active association with Visiting Professors Jianqing Fan and Peter Hall and Emeritus Professors Jim Durbin and Howell Tong. Postdoctoral research staff include Dr Haeran Cho|, who has recently completed her PhD at LSE.
The group also has a strong link with the Econometrics group in the Economics Department, which includes Professors Peter Robinson|, Javier Hidalgo and other eminent time series analysts.
Research interests in the group encompass many aspects of the discipline. Methodological research is guided by applications with the aid of both academic and industrial experts. The theoretical activity in recent years covers a wide spectrum, ranging from linearity to nonlinearity, from parametric to nonparametric and semiparametric, from the stochastic to the deterministic, from spectral approach to wavelets, from spatial to space-time modelling. Applications include biological, economic and financial. e.g. data-analytic approach for inference for volatility of financial time series, structural modelling of weather series and etc. Recent consulting work includes forecasting passport demand for the UK Identity and Passport Service and bootstrapping changes in performance regimes for Maddox Consulting Ltd.
The Centre for the Analysis of Time Series (CATS)|, which is affiliated with the Department, is at the frontier of multidisciplinary research beyond the expertise of any single institution. The Centre works with EPSRC, NERC, the european Commission and directly with industry funding to continue in tracing the uncertainty in weather and climate forecasts from its origins in observations and models error to statements on the reliability of existing forecast systems and available measures of likely economic impact. The Meteorological Office and the European Centre for Medium Range Forecasts are enabling partners in this research: industrial partners include Munich Re, Lloyds of London, Unilever, NationalGrid UK and EDF Energy.
Please visit the Centre for the Analysis of Times Series (CATS)| website for further information about the Centre's research.
Research papers
A small selection of recent reseach papers
Adaptively varying coefficient spatio-temporal models|. Lu, Z., Steinskog, D. J., Tjøstheim, D., and Yao, Q. J. Roy Statist. Soc. B, 2009, 71, 859-880
Sparsistency and rates of convergence in large covariance matrix estimation|. Lam, C., and Fan, J. 2009. Annals of Statistics, 37, 4254-4278
Improved penalization for determining the number of factors in approximate static factor models|. Alessi, L., Barigozzi, M., and Capasso, M. 2010. Statistics in Probability Letters, 2010, 80
Multiscale and multilevel technique for consistent segmentation of nonstationary time series|. Cho, H., and Fryzlewicz, P. 2010. To appear in Statistica Sinica
Inference for stochastic volatility models using time change transformations|. Kalogeropoulos, K., Roberts, G. O., and Dellaportas, P. 2010. Annals of Statistics, 38 (2), pp. 784-807.
Please see the staff pages or LSE Research Online| for a comprehensive list of research outputs.
Research grants
For a full list of all current and recent research grants for the Department of Statistics and the Centre for the Analysis of Time Series (CATS) please see Research Grants|.
For a list of CATS research grants please see here|.
Current MPhil/PhD students
Baojun Dou
Research title: to be confirmed
Na Huang
Research title: New statistical methods for the analysis of high-dimensional high-frequency financial data
Karolos Konstantinos Korkas
|Research title: Nonstationary multivariate times series
Roy Rosemarion
|Research title: Dimension reduction for probability density function estimation
Nikolaos Sgouropoulos
Research title: High frequency, high dimensional volatility modelling and forecasting
Billy Wu
Research title: to be confirmed
Yang Yan
Research title: Quantile regression methods of co value-at-risk
For a full list of all current doctoral research students in the Department of Statistics and Centre for the Analysis of Times Series (CATS) please see Research Students|.
For a list of CATS doctoral research students please see here|.
News
Financial time series analysis workshops and presentations, June 2012
A series of workshops and presentations on financial time series analysis: high-dimensionaliy, non-stationarity and the financial crisis take place at the Institute of Mathematical Sciences, National University of Singapore, on 1-22 June 2012. Professors Piotr Fryzlewicz and Qiwei Yao are co-chairs of the organising committee, alongside Professor Ying Chen of NUS.
The organising committee has issued a call for graduate student posters. The submission deadline is 15 May 2012.
Further details can be found on the conference website|.