Senior Research Associate at Lancaster University, UK

The International Society for Bayesian Analysis; Bailrigg; Presencial; GBP 32.958 - 38.183; Jornada completa

Descripción de la vacante

A leading academic institution in Bailrigg, UK, seeks a Senior Research Associate to develop methodologies in non-reversible MCMC. The researcher will collaborate with teams from the National University of Singapore and the University of Glasgow to enhance computational statistics. Strong programming skills and an understanding of traditional MCMC algorithms are critical. This position offers a competitive salary and a collaborative research environment.

Formación

• Good understanding of standard, reversible MCMC is essential.
• Proven proficiency in computer programming required.

Responsabilidades

• Develop new methodologies in non-reversible MCMC.
• Collaborate with Dr. Chris Sherlock and teams from other universities.
• Implement algorithms in an easy-to-use package.

Conocimientos

Proficiency in computer programming

Understanding of the Metropolis Hastings algorithm

Descripción del empleo

Senior Research Associate at Lancaster University, UK

Aug 3, 2017

Senior Research Associate in Non-Reversible MCMC

Mathematics & Statistics, Lancaster University
Salary: £32,958 to £38,183
Closing Date: Monday 18 September 2017
Interview Date: To be confirmed
Reference: A1924
To apply: https://hr-jobs.lancs.ac.uk/Vacancy.aspx?ref=A1924

Informal enquires to be addressed to Dr Chris Sherlock ([e-mail]).

A Senior Research Associate position for up to three years dedicated to developing new methodologies in non-reversible Markov chain Monte Carlo (MCMC) and led by Dr Chris Sherlock is now available. In collaboration with Dr Sherlock and members of the team at the National University of Singapore and the University of Glasgow, the postdoctoral researcher will develop new, efficient non-reversible algorithms, test and analyse them, both by simulation and theoretically, and implement them in an easy-to use package.

Most standard MCMC algorithms, such as the Metropolis Hastings algorithm, are reversible. However, it is now well established that non-reversible MCMC algorithms can have substantially better mixing properties, particularly for the high-dimensional and complex models that are common in modern applications. Developing general purpose non-reversible MCMC algorithms is currently one of the most active areas of computational statistics.

A good understanding of standard, reversible MCMC is essential, as is a proven proficiency in computer programming.

Starting date: 1 November 2017 or a later date by arrangement.

We welcome applications from people in all diversity groups.