Angus Lewis

Current interests include:

• The application and theory of matrix-analytic methods and matrix-exponential distributions. In particular, using the probabilistic or physical interpretations of processes to analyse them.
• Applied, near real-time tracking and optimal track association using diverse infromation sources and formats, and using methods such as stochastic filtering, statistical parameter matching, clustering, ranking, optimisation.
• Applied statistics for stochastic processes such as hidden Markov time-series models, signal processing and signal detection, new methods for sclerochronology (ecological time-series), estimating parameters of continuous-time Markov chains and branching processes.
• Approximations of stochastic processes such as multi-type branching processes with applications to fitting them to data, or the approximation of fluid queues with applications to numerical methods for more complex processes.
• Optimisation, stochastic optimisation and Markov decision processes.