Predictive Modelling of Stock Option Price Movements Using Machine Learning Techniques: An Indian Market Perspective

The Indian derivatives market, particularly options trading, has witnessed exponential growth over the past decade. However, the nonlinear and volatile nature of option price movements presents a significant challenge for traders, analysts, and financial institutions. This paper aims to develop a predictive modeling framework using machine learning techniques to forecast stock option price trends in the Indian market. Utilizing historical data from the National Stock Exchange (NSE) and Yahoo Finance for major indices like Nifty and Bank Nifty, the study considers factors such as option Greeks (delta, gamma, theta, vega), implied volatility, strike price, and time to expiry. Several supervised learning models including Random Forest, Gradient Boosting, and XGBoost are employed to classify future option price direction (up or down) and to predict price levels. The performance of these models is evaluated using accuracy, precision, recall, F1-score, and RMSE metrics. The results demonstrate that ensemble-based models significantly outperform traditional approaches, with implied volatility, delta, and underlying asset price being key predictors. This research highlights the practical applicability of machine learning in financial derivatives and offers a data-driven approach for improving decision-making in options trading.