Seasonal Volatility Models

• Activity Number: 423
• Type: Contributed
• Date/Time: Tuesday, August 3, 2010 : 2:00 PM to 3:50 PM
• Sponsor: Business and Economic Statistics Section
• Author(s): Ankit Doshi*+ and Julieta Frank* and Aerambamoorthy Thavaneswaran
• Companies: University of Manitoba and University of Manitoba and University of Manitoba
• Keywords: Forecast error ; GARCH ; Kurtosis ; Seasonality ; Volatility

Abstract

Much research has found seasonal behavior in the volatility of financial and commodity data. Most of it has used GARCH models with dummy variables in the volatility equation to account for specific events. A few studies have extended these models to more flexible seasonal forms such as the periodic GARCH. Here, we introduce the multiplicative seasonal GARCH model, where autocorrelation exists at seasonal and at adjacent non-seasonal lags. Building on previous research suggesting that returns series may not be conditionally normally distributed but rather conditionally leptokurtic, we derive the expression for the kurtosis of the multiplicative seasonal GARCH process and for the variance of the l-steps-ahead forecast error.