Comparative analysis of machine learning and deep learning techniques for prediction of the stock market

Abstract

In recent years, there has been a continuing search for reliable instruments that can predict trends in financial markets and activities related to investments. In the past, academics have used traditional methods to forecast the investment worth of equities by analyzing metrics such as the financial records of companies from both a fundamental and technical point of view. The effectiveness of these strategies could decrease as market information asymmetry continues to rise and high-frequency trading becomes increasingly prevalent. Researchers have developed novel methodologies as a result of the progress that has been made in the field of artificial intelligence technology. One of these methodologies is the application of neural networks for forecasting. In the meantime, data visualization is becoming increasingly common, which could make it easier to conduct an in-depth analysis of the advantages and disadvantages presented by various models. The purpose of this research is to evaluate the performance of machine learning and deep learning strategies, including logistic regression, support vector machine, multi-layer perceptron and convolution neural networks, in forecasting stock market prices where various data visualization techniques are utilized for investigation. The findings from error analysis demonstrate that convolutional neural networks operate superbly.