This document discusses using machine learning techniques to forecast financial time series and predict stock market prices. It provides an overview of various machine learning and statistical methods that have been used for stock prediction, including regression, support vector machines, decision trees, neural networks, and random forests. The authors aim to formulate short-term and long-term predictions of stock price direction, changes in price, and actual price. They collect historical stock price and technical indicator data and use feature selection and scaling before applying classification and regression models to achieve 81% accuracy for trend direction and RMSE errors of 0.0117 and 0.0613 for next day price and change, respectively.