Composite Likelihood

Supervisor contact

Related areas

Composite likelihood is a method for fitting and doing inference with models that is used when the full likelihood is computationally too intensive and/or even intractable. This project will review the current literature on composite likelihood methods, with particular focus on doing inference such as hypothesis testing and variable selection. An empirical comparison of composite versus full likelihood methods for particular examples is also possible.

This project will be co-supervised with Dr Francis Hui.

Updated:  23 October 2017/Responsible Officer:  Director/Page Contact:  School Manager