Murray Pollock

Brief Biography

I am a PhD student working under the supervision of Prof. Gareth Roberts and Dr. Adam Johansen in the Department of Statistics at the University of Warwick.

Prior to coming to Warwick I worked as an actuary for Cigna Corporation, initially based in their global HQ in Philadelphia, USA, where I was the reserving actuary for the US domestic life and accident business, and subsequently based in the regional HQ in Madrid, Spain, where I was the lead modeller and developer for the European Region. I also spent a suitably brief time as a trainee working for Mercer HR in their Edinburgh, UK retirement practice.

Current Research

At the moment I am developing computationally efficient methods to simulate (without approximation error) finite dimensional representations of (jump) diffusion sample paths. Further research considers the simulation of upper and lower convergent dominating processes which enfold almost surely (jump) diffusion sample paths over some finite interval. Other research includes extending methodology for particle filters with (jump) diffusion latent transition density.

In general I have interest in:-

Monte Carlo methods (particularly SMC and MCMC)
Computational methods for Stochastic Differential Equations (particularly Exact Simulation)
Computational Statistics (including inference for intractible models)

Publications & Pre-Prints

M. Pollock, A. M. Johansen and G. O. Roberts, On the Exact and $\epsilon$ -Strong Simulation of (Jump) Diffusions. ArXiv mathematics e-print 1302.6964, February, 2013. [arXiv]
M. Pollock, A. M. Johansen and G. O. Roberts, A bounded finite variance Poisson estimator, with application to particle filters for partially observed (jump) diffusions. In preparation.