Stock Exchange Price Prediction Using Linear Regression Model

Abstract

Stock markets are more important to institutions where daily trades. In these trades, millions and billions of dollars are involved. Most of the peoples are investing as a quick way to boost their bank accounts, while others take a more conventional strategy to do so in order to gain long-term rewards. The primary objective of this paper is to create a pattern from past data in order to give an approximate predicting output and a general idea of future values. Several areas, including risk management, consumer data overview, fraud prevention, and stock market projections are use machine learning algorithms. Algorithms for anticipating market values and trends are much more well-known than before in the time of immense and dynamic data. The main aspect of accurate forecasting of stock prices is a high level of detail. We gathered data from the Yahoo Stock Market for two years and offered a detailed customization of feature engineering and machine learning-based models for forecasting stock market price trends. Financial data such as stock open, high, low, and close rates are used as inputs to the model. Many types of regression models are their but in this document are implemented Linear Regression Model using straight forward technique to predict the stock price of the commodity to find the future 5 days closing price of the stocks. The proposed Linear regression (LR) gives the highest performance with an accuracy of 92% than existing model.