In this seminar Beili Zhu will provide an overview of her recent paper, Bayesian analysis of a moving average stochastic volatility model with leverage and heavy-tailed distributions using scale mixtures.

This paper introduces a moving average stochastic volatility model with leverage and heavy-tailed distribution using the scale mixture of normal distributions. In terms of the parameter estimation, an efficient method based on Markov chain Monte Carlo (MCMC) algorithm is developed in this paper. In addition, Beili compares the new model with seven existing stochastic volatility models for their statistical properties using simulated data and three types of real data series. More specifically, the weekly return data of NY Harbor No. 2 Heating Oil and NY Harbor Conventional Gasoline Regular and the daily return data of Equity Hedge are analysed by the author. The model selection criteria relies on the logarithm of the marginal likelihood. The empirical results reveal that the proposed model provides better model fitting than other stochastic volatility alternatives for the returns of the NY harbor No.2 heating oil and Equity Hedge and can beat some of the stochastic volatility models when dealing with the weekly return of the US Gulf Coast Conventional Gasoline Regular.

Beili Zhu is a PhD student scholar at CAMA in Crawford School of Public Policy. Her current research focuses on Bayesian statistics with energy prices forecasting, energy prices modeling and macroeconomics applications of energy prices.

The CAMA Macroeconomics Brown Bag Seminars offer CAMA speakers, in particular PhD students, an opportunity to present their work in progress in front of their peers, and reputable visitors to showcase their work.