Hao Wang

Department of Statistical Science

Duke University


Bayesian Multivariate Modelling: Graphical Models and Time Series

Modelling and inference with higher-dimensional variables, including studies in multivariate time series analysis, raise challenges to our ability to "scale-up" statistical approaches that involve both modelling and computational issues. Modelling issues relate to the interest in parsimony of parametrization and control over proliferation of parameters; computational issues relate to the basic challenges to the efficiency of statistical computation (simulation and optimization) with increasingly high-dimensional and structured models. I am interested in these questions and in Bayesian approaches inducing relevant sparsity and structure into parameter spaces, with a particular focus on time series and dynamic modelling.

I will outline some of my research on integrating statistical graphical modelling ideas into multi- and matrix-variate dynamic models of various kinds. This will include the development and application of dynamic graphical models for multivariate financial time series and matrix-variate econometric examples, and dynamic multivariate regression modelling with simultaneous selection of variables and graphical-model structured covariance matrices. The talk will highlight a blend of Bayesian modelling and computational developments as well as examples, and mention potential extensions, additional areas for application, and a number of new and open research directions.


Back to Colloquium Series