Speaker: Richard Gerlach
Date: Friday, June 15, 2018
Venue: Room 335, HSBC Business School Building
Richard Gerlach is a Professor of Business Analytics at Business School, the University of Sydney. His research interests lie mainly in financial econometrics and time series. His work has concerned developing time series models for measuring, forecasting and managing risk in financial markets as well as computationally intensive Bayesian methods for inference, diagnosis, forecasting and model comparison for these models. Recent focus has been on nonlinear threshold heteroskedastic models for volatility, Value-at-Risk and Expected Shortfall forecasting. He has developed structural break and intervention detection tools for use in state space models; also has an interest in estimating logit models incorporating misclassification and variable selection. His research papers have been published in Journal of the American Statistical Association, Journal of Business and Economic Statistics, Journal of Time Series Analysis and the International Journal of Forecasting. He has been an invited speaker and regular presenter at international conferences such as the International conference for Computational and Financial Econometrics, the International Symposium on Forecasting and the International Statistical Institute sessions.
Inversion copulas show promise in modelling latent nonlinear state space models with Markov dependence structures. We extend this idea to cover nonlinear time series with non-Markov dependence, with focus on two special cases: the well-known GARCH and Realized GARCH specifications. Both present challenges in finding and evaluating the implied margin of the latent variable: we discuss some possible solutions here. Likelihood and Bayesian computational methods are derived for estimation, inference and forecasting purposes. The proposed time series inversion copula models are used to model and forecast financial returns from several financial indices, including an emerging markets index and a gold and silver index. The proposed models are competitive for density and tail risk forecasting in these series, compared to a range of popular, competing financial time series models.