Stock Market Price Prediction Using Machine Learning Algorithms for Investment Decision-Making

Abstract

This study, titled "Stock Market Price Prediction Using Machine Learning Algorithms for Investment Decision-Making," evaluates the economic viability and operational efficiency of implementing algorithmic prediction models in retail and institutional portfolio management. In modern financial markets, predicting stock prices is highly challenging due to nonstationarity, noise, and complex nonlinear relationships. This research investigates the implementation of machine learning algorithms—specifically Linear Regression, Support Vector Regression (SVR), Random Forest, and Long Short-Term Memory (LSTM) Networks—for real-time price forecasting. The study conducts a cost-benefit analysis of the technological investment, evaluating capital expenditure, operational maintenance costs, and the net financial benefit of stock price forecasting and portfolio optimization from 2021 to 2025. Standard financial appraisal metrics—Net Present Value (NPV), Internal Rate of Return (IRR), Payback Period (PBP), and Benefit-Cost Ratio (BCR)—are applied to determine the long-term profitability of the investment. The empirical analysis indicates that the LSTM model achieved the highest prediction accuracy, with a Mean Absolute Percentage Error (MAPE) of only 2.1%. An ML-optimized portfolio generated consistent excess returns (alpha) over the Nifty 50 Index across the five-year planning horizon, yielding a positive NPV of 392.4 Crores and an IRR of 45.1%. The study concludes that the integration of machine learning algorithms into investment decision-making processes represents a highly viable and financially feasible strategy for modern banking and asset management portfolios, delivering significant risk-adjusted financial returns.