Model specification and news announcements: theoretical and empirical aspects of option pricing

This thesis covers the e?ects of macroeconomic announcements and model specification on option pricing. In the first main chapter a novel design for an approximative solution to a non-affine stochastic volatility model for option pricing is derived, inspired by methods frequently used in physics. This methodology is then compared against numerical solutions and is found to perform very well. In the second main chapter we conduct an empirical study of high-frequency option pricing on S&P 500 options. The chapter has two objectives: to examine the equity price uncertainty surrounding information released in FOMC announce-ments. Additionally, we quantify this impact on formal option pricing models. Secondly, to investigate option market maker biases to volatility accrual and this biases ability to explain why option returns, on average, are negative overnight and positive intraday. We also investigate this e?ect on formal option pricing models. In the third main chapter we investigate equity option returns and their behaviour around earnings announcements. We document a robust novel finding that delta-hedged equity option returns increase with increasing earnings announcement variance of the underlying stock. This result is separate from existing anomalies in stock and options markets and cannot be explained by standard risk factors. The thoughts demonstrated in this thesis lay the foundations for further interesting research and practically relevant applications. Please note that the first main chapter has been publish in the journal of Futures Markets, DOI: fut.22061.