Title: Bayesian modeling of high frequency stochastic volatility with intraday seasonality and skew heavy-tailed error
Authors: Makoto Nakakita - Keio University (Japan) [presenting]
Teruo Nakatsuma - Keio University (Japan)
Abstract: Intraday high frequency data of asset returns exhibit not only typical characteristics (e.g., volatility clustering) but also a cyclical pattern of return volatility that is known as intraday seasonality. We extend the stochastic volatility (SV) model for application with such intraday high frequency data and develop an efficient Markov chain Monte Carlo sampling algorithm for Bayesian inference of the proposed model. Our modeling strategy is two-fold. First, we model the intraday seasonality in return volatility with orthogonal polynomials and estimate it along with the stochastic volatility simultaneously. Second, we incorporate a possibly skew and heavy-tailed error distribution into the SV model by assuming that the error distribution belongs to a family of generalized hyperbolic distributions such as variance-gamma, Student's $t$ and their skew variants. As a demonstration of our new method, we estimate the proposed model with 1-minute and 5-minute return data of a stock index (TOPIX) and conduct comparison among competing model specifications with the widely applicable information criterion (WAIC). The results show that the SV model with skew variance-gamma error is the best in a volatile market.