A Quantitative Model for Option Sell-Side Trading with Stop-Loss Mechanism by Using Random Forest

Abstract

Due to the characteristics of high leverage and low margin, option is very suitable for quantitative trading by applying portfolio management to control the profit and risk. The money management is an important issue to build a portfolio especially for option sell-side trader, since the profit is only the premium, while the loss is unlimited. In this research, we propose a model for option sell-side strategy to estimate the win-rate of option by the premium, time to maturity, and volatility based on statistical approach and random forest algorithm. The prediction of the model is visualized through heatmap which can reveal the profitable trading range intuitively, we use the precision score to evaluate the performance in these two models and proof the effectiveness and robustness of predictive model proposed by random forest algorithm. In the future, we plan to apply other machine learning algorithm to propose the predictive model for spread trading.