Stock market prediction based on adaptive training algorithm in machine learning

Abstract

This study deals with one of the most important issues for understanding financial markets, future asset fluctuations. Predicting the direction of asset fluctuations accurately is very difficult due to the uncertainty of the stock market, the influence of various economic indicators, and the sentiment of investors, etc. In this study, we present a new method to improve the effectiveness of machine learning by selecting appropriate training data using an adaptive method. The application to various sector data of the S&P 500 and many machine learning methods shows that the proposed adaptive data selection algorithm improves the prediction accuracy of the stock price direction. In addition, it can be seen that the adaptive data selection method increases the return on the asset investment.

Keywords:

• Adaptive data construction
• Empirical validation
• Financial forecasting
• Machine learning

JEL Classifications:

• C53
• C63
• C18

Disclosure statement

No potential conflict of interest was reported by the authors.

Additional information

Funding

This work was supported by the National Research Foundation of Korea (NRF) funded by the Korean government (No. 2021R1F1A1054766 and NRF-2018R1D1A1B07050046).