Professor of Statistics, Professor Jim Griffin, from the School of Mathematics, Statistics and Actuarial Science (SMSAS) at the University of Kent, gave a talk at an Econometrics Seminar in the Netherlands in October.
Jim presented a talk titled, ‘Bayesian Nonparametric Vector Autoregressive Models’ based on collaborative research with Lecturer in Statistics, Dr Maria Kalli, at the Department of Econometrics at the University of Rotterdam.
Abstract:
Vector autoregressions (VARs) are the main method used by macroeconomists for modelling multivariate, economic time series data. The models are simple to interpret and fit but their assumptions of stationarity, linearity and normality are often too restrictive. In this talk, I will consider developing a Bayesian nonparametric VAR model which address some of these shortcoming and allows us to identify different regimes in the data. The model will be illustrated and compared to competing models on monetary policy and macro-financial linkage examples.
Professor Jim Griffin
Professor in Statistics
Jim is Deputy Head of School and a Professor of Statistics. Jim’s research interests include: Bayesian nonparametric methods, the application of nonparametric and semiparametric methods to economic and financial data, and sparse Bayesian methods for regression problems with many variables.
.
Dr Maria Kalli
Lecturer in Statistics
Maria is a Lecturer in Statistics and joined the School in 2017. Her research interests include: Econometrics, mathematical finance, Bayesian non-parametric methods, Bayesian regression and variable selection, MCMC methodology, and financial time series modelling.
.