A Comprehensive Review On Stock Market Price Forecasting Approaches

Abstract

The stock market is a crucial aspect of a nation's economic and social structure. Forecasting stock prices in the stock market presents a rigorous and very challenging task to investors, professional analysts, and market researchers in the financial market because of stock price time series, which are noisy, nonparametric, volatile, complex, nonlinear, dynamic, and chaotic. Stock market prediction is a vital problem and an eminent research topic in the financial domain because investment in the stock market has more risk. Nonetheless, applying computational intelligence techniques has made it possible to minimize this risk considerably. This paper is devoted to the approach of using the methods of computational intelligence to forecast the stock market, including the use of Machine Learning (ML) and Deep Learning (DL) techniques to forecast. A comprehensive study of stock market price forecasting is presented based on computational intelligence methods. Additionally, it summarizes the advantages, drawbacks, datasets used, and performance metrics of these algorithms in tabular form. Finally, it addresses the challenges in stock market price forecasting and recommends promising solutions to boost prediction accuracy.