Michael Smith joined MBS in 2007. Prior to that he was an Associate Professor of Econometrics at the University of Sydney and a Senior Lecturer of Econometrics at Monash University. He has held visiting positions at the University of Munich, The Wharton School at the University of Pennsylvania and the McCombs School of Business at the University of Texas. Michael's research involves the development of Bayesian estimation methods, along with econometric modelling and forecasting. He is prominent internationally for his work on Bayesian semiparametric smoothing and model averaging in cross-sectional, spatial and time series contexts. His more recent work has involved the development of methods for the analysis of multivariate dependence using copula models. His research has appeared widely in the top academic journals in econometrics and statistics. He has taught courses in econometrics, statistics and quantitative finance at undergraduate and postgraduate levels, including specialised PhD level courses. At MBS he teaches into the MBA and Executive MBA programs. He also has a long-standing interest in econometric problems arising in the energy sector, including the modeling and forecasting of both demand and spot prices within deregulated wholesale electricity markets.
Forthcoming and Working Papers
A Comparison of Periodic Autoregressive and Dynamic Factor Models in Intraday Energy Demand Forecasting (with Thomas Mestekemper and Goeran Kauermann), Forthcoming in International Journal of Forecasting (2012)
We suggest a new approach for forecasting energy demand at an intraday resolution. Demand in...
Estimation of Copula Models with Discrete Margins via Bayesian Data Augmentation (with Mohamad A. Khaled), Forthcoming in Journal of the American Statistical Association (Theory and Methods) (2011)
Estimation of copula models with discrete margins is known to be difficult beyond the bivariate...
Public Datasets & Software
Windows Executable for Gaussian copula with NBD margins (2011)
This is an example Windows 32bit program to estimate a Gaussian copula model with NBD...
Articles
Modeling Dependence using Skew t Copulas: Bayesian Inference and Applications (with Quan Gan and Robert Kohn), Journal of Applied Econometrics (2012)
[THIS IS AN AUGUST 2010 REVISION THAT REPLACES ALL PREVIOUS VERSIONS.]
We construct a copula...
Bicycle Commuting in Melbourne During the 2000s Energy Crisis: A Semiparametric Analysis of Intraday Volumes (with Goeran Kauermann), Transportation Research Part B: Methodology (2011)
Cycling is attracting renewed attention as a mode of transport in western urban environments, yet...
Forecasting Television Ratings (with Peter Danaher and Tracey Dagger), International Journal of Forecasting (2011)
Despite the state of flux in media today, television remains the dominant player globally for...
Modeling Multivariate Distributions Using Copulas: Applications in Marketing (with Peter J. Danaher), Marketing Science (2011)
In this research we introduce a new class of multivariate probability models to the marketing...
Rejoinder: Estimation Issues for Copulas Applied to Marketing Data (with Peter J. Danaher), Marketing Science (2011)
Estimating copula models using Bayesian methods presents some subtle challenges, ranging from specification of the...
Contributions to Books
Bayesian Approaches to Copula Modelling, Working Paper (2011)
Copula models have become one of the most widely used tools in the applied modelling...
Bayesian Inference for a Periodic Stochastic Volatility Model of Intraday Electricity Prices, Statistical Modelling and Regression Structures: Festschrift in Honour of Ludwig Fahrmeir (2010)
The Gaussian stochastic volatility model is extended to allow for periodic autoregressions (PAR) in both...