Skip to content
Open access

Machine Learning-Based Stock Market Volatility Prediction for Intelligent Investment and Financial Risk Management

Sep 2026 · Computer and Decision Making: An International Journal · 0 citations

Abstract

Stock market volatility represents a major source of uncertainty for investors and financial institutions, creating a need for reliable predictive tools that can support risk-aware decision-making. This study developed a machine learning framework for forecasting future Tesla Inc. stock volatility using historical daily market data and engineered financial indicators. The dataset contained 3,499 observations, from which returns, moving averages, lagged variables, technical indicators, and 7-day and 20-day historical volatility measures were derived. Future 7-day realised volatility was used as the prediction target. Four models—Random Forest Regression, Support Vector Regression, XGBoost Regression, and Long Short-Term Memory were evaluated using a chronological 80:20 train-test split. LSTM achieved the best overall performance with an RMSE of 0.2128 and R² of 0.0799, while Random Forest produced the lowest MAE of 0.1729. Twenty-day historical volatility emerged as the most influential predictor, indicating persistence in market volatility. Risk classification showed that 53.6% of test observations were high risk, and LSTM correctly identified 82.8% of actual high-risk cases. The findings demonstrate that machine learning can provide useful but modest volatility forecasts and can function effectively as a supplementary tool for intelligent investment and financial risk management.

Read PDF

We use cookies to run the site and, with your consent, for analytics and to show ads. See our Cookie Policy.