Machine Learning Techniques for Stock Market Prediction

This work provides an in-depth evaluation of three prominent machine learning approaches utilized in the prediction of stock market movements. It covers: (1) a hybrid model combining ARIMA with Support Vector Machines to address both linear and non-linear components of stock price series, (2) a two-stage architecture that integrates Support Vector Regression with Artificial Neural Networks and Random Forests to enhance prediction stability, and (3) an application of Long Short-Term Memory (LSTM) networks for directional forecasting based on temporal dependencies in financial data. Each technique is analyzed in terms of input features, experimental design, forecasting horizons, performance indicators, and real-world relevance. The review aims to give undergraduate students, early-stage researchers, and data science enthusiasts a clear and comparative understanding of current strategies used in financial prediction through machine learning. Keywords: stock price forecasting, machine learning, ARIMA, SVM, LSTM, financial modeling, time series prediction, ensemble learning, review study