Seasonal Stochastic Volatility Models

• Activity Number: 360
• Type: Contributed
• Date/Time: Tuesday, August 2, 2011 : 10:30 AM to 12:20 PM
• Sponsor: Business and Economic Statistics Section
• Author(s): Julieta Frank*+ and Melody Ghahramani and Aerambamoorthy Thavaneswaran
• Companies: University of Manitoba and University of Winnipeg and University of Manitoba
• Keywords: kurtosis ; forecast error variance ; random coefficient ; seasonality ; stochastic volatility

Abstract

New information entering into the market causes changes in price volatility. Often these changes occur at regular time intervals, for example, at market opening and closing times, at certain times during the day, at specific days of the week, or around weekends and vacation periods. Even though much research has been done on volatility models, more general specifications accounting for seasonal volatility have been little explored. In this paper, we focus on the random coefficient autoregressive (RCA) models with seasonal stochastic volatility (SV) errors. We derive the moments of the RCA-seasonal SV model and then we obtain the closed-form expression for the variance of the l-steps ahead forecast error. The results are a generalization of the non-seasonal version of the model. The expressions derived here can be used to obtain the moment estimates of the model parameters to provide better forecasts of call prices and market behavior. The results may be useful to investors, decision makers, and other market participants for the development of trading strategies.