Date Thesis Awarded


Access Type

Honors Thesis -- Access Restricted On-Campus Only

Degree Name

Bachelors of Science (BS)




Ed Chadraa

Committee Members

Haipeng Chen

Greg Hunt


Volatility in the stock market has increasingly become a target of investigation. Understanding and predicting volatility in the stock market can give one great advantages in investment options. Over the years, many models have been developed, such as the Autoregressive Moving Average (ARMA) and the Autoregressive Conditional Heteroskedasticity (ARCH) models. The ARCH model has served as the foundation for other models like the General ARCH (GARCH) and the Continuous GARCH (COGARCH) models. In this paper, we investigate the effectiveness of the widely used GARCH(1,2) model to the relatively new COGARCH(1,2) model that was conceptualized by Brockwell, Chadraa, and Lindner, in predicting volatility in S&P500 index, namely the VIX index. Specifically, we look at each model’s one step prediction directional accuracy and percent errors in both models.

On-Campus Access Only