When: Thursday 5th July 2018, 13:00-14:00.
Included in meeting:
The fortnightly BRAG meeting will be held 05/07/18 at 1pm in room Y801. This week we will have presentations by Dan and Mahasen.
Dan’s Talk: Efficient Bayesian Estimation for Bad Environment – Good Environment (BEGE) Model via SMC
Abstract : Sequential Monte Carlo (SMC) is an efficient method for sampling from the posterior distribution of static Bayesian models and can be regarded as an alternative to Markov chain Monte Carlo (MCMC) methods. SMC may be preferred over MCMC for certain classes of problems, since SMC is easily parallelisable, adaptable and more effective in sampling from irregular or complex posterior distributions. Despite the advantages of SMC being widely known to statisticians, the methods have not been adopted and studied in some applied fields. In this talk, I will present a novel application of SMC to an econometric model – bad environment – good environment (BEGE) model. We aim to develop a fully Bayesian approach for the BEGE model and used for estimation, prediction and model selection.
Mahasen’sTalk: Synthetic likelihood using synlik R package
Abstract: Models used in epidemiology and ecology often have complex likelihood functions which make computationally infeasible or impossible to undertake inference. Synthetic likelihood is an alternative method to approximate likelihood function when it is possible to simulate data from those complex models. In this talk, I will give a brief introduction to undertake bayesian inference for models with intractable likelihoods using “synlik” R package.
We look forward to seeing you all there!
Farzana and Trish