Abstract: Accurate prediction of stock market values is a critical task in financial analysis, empowering investors to make informed decisions. Machine learning has emerged as a powerful approach to enhance the authenticity and effectiveness of stock market forecasting. This research paper focuses on investigating the potential of regression models and LSTM-based machine learning techniques for predicting stock values. By comparing the performance of these models in stock market valuation, we aim to uncover their strengths and limitations. Our study leverages comprehensive historical stock market data from diverse sources, which undergoes meticulous preprocessing to extract pertinent features such as price trends, trading volume, and market sentiment. Regression models such as linear regression, polynomial regression, and support vector regression are implemented and rigorously evaluated to assess their predictive capabilities in estimating stock prices accurately. Additionally, we explore the potential of LSTM-based deep learning models in capturing intricate temporal dependencies and patterns in the data.
Keywords: stock market , forecasting, price prediction ,machine learning.
|
DOI:
10.17148/IJARCCE.2023.12645
