Abstract

In this thesis, I empirically compare the pricing performance of three classes of stochastic volatility option pricing models and the traditional Black-Scholes (1973) model in the pricing of S&P Canada 60 Index Options. The stochastic volatility models that I study are as follows: (1) the ad hoc Black and Scholes (1973) procedure that fits the implied volatility surface, (2) Madan et al.'s (1998) variance gamma model, and (3) Heston's (1993) continuous-time stochastic volatility model. I find that Heston's continuous-time stochastic volatility model outperforms the other models in terms of in-sample pricing and out-of-sample pricing. Second, the addition of the stochastic volatility term to the stochastic volatility model and variance gamma model does not resolve the "volatility smiles" effects, but it reduces the effects. Third, the Black-Scholes model performs adequately in pricing options, with the advantage of simplicity, although it suffers from the shortcoming of the "volatility smiles" effect. Finally, although it includes more parameters, the ad hoc Black and Scholes model does not perform as well as expected.