Analysis of the Difference in Stock Price Between A-shares and American Stocks in Machine Learning

¹ School of accounting, Guangdong Baiyun University, 519000 Guangzhou, China
² School of international Education, Henan University of Animal Husbandry and Economy, 450000 Zhengzhou, China

^* Corresponding author: Yuhao.Liu@calhoun.edu
^† These authors contributed equally.

Abstract

Contemporarily, stock market is the most representative financial investment tool in the world. The application of machine learning has had a significant impact on the development of society and economy as well as productivity, and has also been inextricably linked to the securities market. This study will analyse and compare the technological development of machine learning in the last five years, as well as the stock value data and stock price fluctuations of A-shares and American stocks in the field of machine learning. In this way, the machine learning technology may change the global stock market in the future, and the prospect of this technology in the future. This paper introduces three forecasting models, namely Light Gradient Boosting Machine (lightGBM) model, Convolutional Neural Networks (CNN) model and Long short-term memory (LSTM) model, and studies their influence on stocks and forecasting accuracy. Applying machine learning to financial investment is a two-edged sword, with advantages and disadvantages, opportunities and challenges, depending on whether and the measure to implement it.