A1009
Title: Quantitative trading of vertical spread option strategies with stop-loss by machine learning
Authors: Min-Kuan Chen - National Taipei University of Technology (Taiwan) [presenting]
Mu-En Wu - National Taipei University of Technology (Taiwan)
Wen-Shuen Wu - National Chengchi University (Taiwan)
Abstract: In recent years, quantitative trading with AI techniques has been developed in finance research and applications. In prior works, quantitative trading studies forecasted the underlying asset dynamics and evaluated the expected value. However, the win rate is hard to explore and predict. We leave the odds by vertical spread option strategies to address the challenge, which could pre-lock profit and loss. Furthermore, we proposed a method to estimate the probability with the stop-loss mechanism as a win rate via a statistical approach and machine learning to improve the performance. The spread strategy position will be closed when the stop-loss is triggered. Otherwise, it will remain open until the option expires. Subsequently, we visualize the win rate by heat-map and select the profitable spread strategy at multiple moments and spreads. The results show that the accumulated profit and loss curve monotonically increases and enhances performance. This suggests that our method generates a promising approach and applies to practical program trading. Other AI techniques, such as neural networks and deep learning, may also predict win rate, as we have demonstrated here using machine learning.
