The fast growth in Artificial Intelligence (AI) and machine learning has brought notable changes to the way financial forecasting and risk management are carried out in global markets. Conventional econometric models, although commonly applied, often face limitations in capturing nonlinear relationships, rapid market fluctuations, and the complex structures present in financial time series data. In this regard, AI-based methods present effective alternatives for improving the accuracy of predictions and supporting better financial decisions. This study explores how artificial intelligence can strengthen financial forecasting and risk management, with a specific focus on Indian financial markets. It compares the effectiveness of traditional econometric approaches—especially the Generalized Autoregressive Conditional Heteroskedasticity (GARCH) model—with modern learning techniques such as Long Short-Term Memory (LSTM) neural networks, which are well-suited for identifying time-dependent patterns in financial data. The analysis is based on secondary data collected from leading financial institutions and stock exchanges in India, including the National Stock Exchange and the Bombay Stock Exchange. To assess forecasting performance, standard evaluation metrics like Mean Squared Error (MSE) and Root Mean Squared Error (RMSE) are used. The results indicate that AI-based models deliver better performance than traditional econometric models in forecasting market volatility and financial risk. This improvement is largely due to their capabilities to handle large data and detect complex nonlinear relationships.
Lokesha N (Thu,) studied this question.