Nick Whiteley
Contact
School of Mathematics
University Walk
Bristol
BS8 1TW
Office 3.5 Main Building
email: nick.whiteley[at]bristol.ac.uk
Research
I am
interested in Monte Carlo methods, especially sequential methods and their application
to a variety of estimation, dynamical inference and optimization problems.
Working Papers
- Twisted
Particle Filters.
With Anthony Lee. 2012. Submitted. [.pdf] and the proofs: [.pdf]
- A particle
method for approximating principal eigen-functions
and related quantities. With Nikolas Kantas. 2012. Submitted.
[arXiv]
- Bayesian
Computational Methods for Inference in Multiple Change-points Models. With Arnaud Doucet and Christophe Andrieu.
2011. Submitted. [.pdf]
- Efficient Bayesian Inference for Switching State-Space
Models using Particle Markov chain Monte Carlo methods. With Arnaud Doucet and Christophe Andrieu.
Bristol Statistics Research Report 10:04. Submitted. [arXiv]
Journal
Publications
- Stability
properties of some particle filters. The Annals of Applied Probability. To appear.
[Preprint]
- Bayesian Learning of Noisy Markov
Decision Processes. With Sumeet Singh and
Nicolas Chopin. ACM TOMACS. To
appear. Linear variance bounds for
particle approximations of time-homogeneous Feynman Kac
formulae. With Nikolas Kantas and Ajay Jasra. Stochastic Processes and their Applications. 2012.
[arXiv]
[Journal]
- Sequential
Monte Carlo samplers: error bounds and insensitivity to initial conditions. Stochastic Analysis
and Applications. To appear. [Preprint]
- Monte Carlo filtering
of piecewise-deterministic processes. With Adam Johansen and Simon Godsill.
Journal of Computational and Graphical Statistics. [Preprint]
- Auxiliary Particle
Implementation of the Probability Hypothesis Density Filter With Sumeet
Singh and Simon Godsill. IEEE Transactions
on Aerospace and Electronic Systems. [Preprint]
Discussion
- Discussion of Particle Markov Chain Monte Carlo
methods by Andrieu, Doucet
and Holenstein, Journal of the Royal Statistical
Society Series B. [.pdf]
Book Chapters
- Recent Developments in Auxiliary Particle Filtering. With Adam Johansen. In
Barber, Cemgil and Chiappa
(editors), Inference
and Learning in Dynamic Models, Cambridge University
Press. To Appear. [.pdf]
- An approximate
likelihood method for estimating the static parameters in multi-target tracking
models.
In Barber, Cemgil and Chiappa
(editors), Inference
and Learning in Dynamic Models, Cambridge University
Press. To Appear. [.pdf]
- Bayesian Statistical Methods for Audio and Music
Processing.
With Taylan Cemgil,
Simon Godsill and Paul Peeling. In O'Hagan and
West (editors), The Oxford Handbook of Applied Bayesian Analysis. Oxford
University Press. To Appear. [.pdf]
Collaborators