When: Thursday 25th May 2017, 13:00-14:00.
Included in meeting:
This week we will have two presentations by Matt and Jagath
“Stan: A probabilistic programming language for Bayesian inference.
Stan is a free open-source C++ program used for arbitrary user specified Bayesian models and can be called from Matlab, Python, R or Julia. I will give a brief introduction from the users’ perspective and illustrate the method with a simple nonlinear regression example. ”
“The Total Entropy Utility Function in Bayesian Sequential Design for Copula Models
This study presents the total entropy utility function for deriving Bayesian designs for the dual purpose of model discrimination and parameter estimation for Copula models. This design approach is useful for designing experiments which yield multiple dependent responses whose multivariate distributions are not available. Here, we explore various Copula models which explain dependence structures in bivariate discrete and mixed responses. The sequential Monte Carlo algorithm is adopted to reduce the computational effort required in deriving efficient sequential designs. Moreover, the performance of the total entropy utility function is evaluated under different Copula models.”
We look forward to seeing you all there!
Matt and Shovanur