An Evaluation of Machine Learning Algorithms for Predicting Closing Stock Prices in the Indian Stock Market

This study examines how effective machine learning algorithms are at forecasting the closing stock prices of five major FMCG companies listed on the National Stock Exchange of India (NSE). The focus is on methods such as Random Forest, Decision Trees, Linear Regression, and Support Vector Regression, aiming to analyze their performance against metrics like accuracy, precision, recall, and F1-score. The goal is to uncover insights regarding the reliability of different machine learning techniques for stock price prediction, ultimately assisting investors and financial analysts in making well-informed investment choices. The study will leverage historical stock price data, incorporating relevant technical indicators and fundamental features to train and test the predictive models. To ensure data quality, a meticulous preprocessing stage will address missing values, normalize features, and minimize potential biases. The models' performance will be evaluated using robust metrics such as Mean Absolute Error (MAE) and R-squared to assess accuracy and overall effectiveness.