Author(s): Tanishq Manoj Jarsodiwala, Parvej Usman Khatik, Shravani Vinod Lakkad, Makarand Shahade,Umakant Mandawkar
The stock market’s volatility and complexity present substantial hurdles for investors seeking to make informed deci- sions.This research proposes a ML-based strategy for predicting stock purchasing decisions using financial variables such as Price to Earnings , Price to Book , Return on Equity , Return on Assets , and Debt to Equity . The project processes and analyzes historical stock data using advanced algorithms such as XGBoost, finding relevant patterns for predictive accuracy. The approach overcomes significant constraints of existing methods by incorporating robust data preprocessing, feature engineering, and model evaluation using metrics such as Mean Absolute Error (MAE) and Root Mean Squared Error (RMSE). A user-friendly interface guarantees that investors may gain actionable insights and make informed decisions.This study not only closes the difference between conventional financial analysis and cutting- edge Machine Learning, but it also provides the framework for scalable applications in financial consulting, education, and real- time stock market forecasting. Future developments will include real-time data processing, portfolio optimization, and expanded feature sets to improve prediction skills.