Development and analysis of derivative trading systems using artificial intelligence.

摘要

This dissertation proposes a methodology that utilizes a generalized regression neural network to develop hybrid option trading systems that incorporate both volatility and return forecasting. This study focuses on the S&P 500 stock index as a representative for the market. The three different trading methods are discussed: stock return forecasting using a simple call and put option strategy, volatility forecasting applying a straddle option strategy, and the combination of volatility and stock return forecasting applying advanced strategies, such as strip, strap, bull, and bear spread strategies. The results show that the hybrid options trading model can improve the overall trading return and outperform trading models using merely return forecasting or volatility forecasting in isolation. A sensitivity analysis for each trading system is investigated to observe the results at different values of option characteristics. The results from the experiment reveal that high delta options should be applied for the long call and long put trading strategies because of their fast response to changes in the price of the underlying asset. In addition, it is difficult to make a profit using daily trading with options due to wide bid-ask spreads in option pricing. In contrast, weekly trading using options can be profitable when utilizing the right combination of parameter adjustments, along with a suitable investment strategy.