This thesis investigates whether there are predictable dynamics in risk-neutral moments extracted from market prices of options on the S&P 500 index. To this end, risk-neutral moments of different horizons are extracted from market option prices. Different econometric models are employed to model the dynamics of the extracted moments. Their forecasting accuracy is assessed in an out-of-sample forecasting study employing both a statistical and an economic setting. The set of best performing forecasting models is detected by employing the newly developed Model Confidence Set approach by Hansen, Lunde, and Nason (Econometrica, 2011). The economic value of risk-neutral moments forecasts is assessed in an option trading case study for which risk-neutral moment trading strategies are devised. It is found that risk-neutral moments can be predicted by a set of models. In particular, autoregressive patterns and cross-moment effects play an important role. Signifcant economic profits can be generated by trading based on risk-neutral skewness forecasts. However, these profits vanish once transaction costs are taken into account. The results have implications for the predictability of implied volatility surfaces, the development of option pricing models, and the efficiency of option markets.
«
This thesis investigates whether there are predictable dynamics in risk-neutral moments extracted from market prices of options on the S&P; 500 index. To this end, risk-neutral moments of different horizons are extracted from market option prices. Different econometric models are employed to model the dynamics of the extracted moments. Their forecasting accuracy is assessed in an out-of-sample forecasting study employing both a statistical and an economic setting. The set of best performing foreca...
»