Plenary Lecture

Linear and Nonlinear Filtering in Mathematical Finance

Dr. Paresh Date
Center for the Analysis of Risk and Optimisation Modelling Applications,
Department of Mathematical Sciences,
Brunel University, UB8 3PH,
U.K.

Email: Paresh.Date@brunel.ac.uk
Website: http://people.brunel.ac.uk/~mastpmd
 

Abstract: The problem of filtering unobservable or latent variables from noisy data arises naturally in many financial applications. This talk will provide a broad overview of time series filtering, with an emphasis on the theory of affine Gaussian filtering (or Kalman filtering) of discrete time series data. Empirical work is presented on two specific filtering applications in finance: modelling the short rate using observed bond yields and modelling the daily volatility of stock price based on the observed intra-day data. The talk concludes with an outline of recently proposed approximate filtering methods for nonlinear time. series.
 

Brief Biography of the Speakers:
Dr Date completed his doctoral studies in systems theory at University at Cambridge, UK. He joined Brunel University, UK as a lecturer in mathematical sciences in 2002, where he is now a senior lecturer. Dr Date has published over 20 refereed papers in the areas of uncertainty modelling, calibration and filtering, with a special emphasis on financial applications. He has held visiting positions in India, Canada and Australia and has given invited talks at several UK and overseas Universities on topics in systems theory. He is an Associate Editor of IMA Journal of Management Mathematics.