EVALUATING MACHINE LEARNING REGRESSION MODELS FOR PREDICTING STOCK MARKET TRENDS FOR THE CONSTRUCTION INDUSTRY: A COMPARATIVE STUDY

Abstract

The accurate anticipation of the stock market index’s trajectory plays a pivotal role in formulating effective share trading methods. The accurate forecasting of closing stock prices has the potential to yield significant advantages for investors. Machine learning algorithms possess the capability to effectively analyze and predict historical stock patterns, hence yielding highly dependable closing price forecasts. This article presents a comprehensive study of the NASDAQ stock market, with a specific focus on constructing a diversified portfolio consisting of ten businesses from the construction industry as well as other industries. The aim of this study is to calculate the opening price of a stock of a construction enterprise or any other business for the following day by utilizing past data. In order to complete the assignment, we employed nine distinct Machine Learning regression models on a dataset consisting of stocks data. The performance of these models was assessed using Mean Squared Error (MSE) and the coefficient of determination (R2).