After completing my MSc(Statistics) from the Indian Institute of Technology Kanpur in 2003, I decided to work as a Statistician in Industry. During my stints with various organisations, I ended up working in the area of Market Research and Econometric Modeling for about a year and on Clinical trials for three years. Finally, however, I was convinced that it wasn't for me (I had had enough!) I decided to go back to academic life and joined the PhD program at the Department of Statistics at Trinity College, Dublin. After completing my PhD, I also did a (short) post-doc at Trinity before joining the Department of Statistics, University of Waikato as a Lecturer in July 2011.
My research interests primarily include
Bayesian modeling of complex systems
The second research interest often stems from the first one. For example, my PhD research was concerned with Bayesian modeling of the dynamic force exerted by vehicles on the road surface - this was essentially a problem of Bayesian inference on the parameters of a system of second order differential equations. However, MCMC based methods implemented on complex systems such as this can be computationally very expensive to the extent that they limit the quality of the inference. Therefore, as part of that research, I developed a new approach to implement computationally efficient Bayesian Inference on Stochastic Differential Equation (SDE) models.
Presently, I am working on writing a proposal to extend this approach to a more general class of models and also explore the ways to improve its efficiency.
I am also interested in the areas of Low-discrepancy sequences and Randomised Quasi-Monte Carlo Integration (RQMC). Currently I am working (along with Prof John Haslett, TCD) on developing a computationally efficient method for estimating the variance of the RQMC estimate. This work is leading to some interesting insights into the nature of RQMC sequences.
Developing computationally efficient (non-MCMC based) methods for inference and
Randomised-Quasi Monte Carlo integration