Option Pricing Using Machine Learning with Intraday Data of TAIEX Option

The use of artificial intelligence (AI) in the financial sector has become increasingly popular in recent years. This study focuses on the application of machine learning (ML) for option pricing using intraday data on Taiwan's Capitalization-Weighted Stock Index (TAIEX). This study compares this method with the traditional Black-Scholes option pricing model to determine if the results of ML are more accurate in predicting market prices. The empirical results show that ML can provide more accurate option pricing than the BS model, particularly when training the model with option volatility. Moreover, the pricing ability of the model is positively correlated with the frequency of data used in this study. However, when predicting prices for the next six months, machine learning does not outperform a BS model using lagged prices.