Abstract
This study aims to predict the direction of US stock prices by integrating time-varying effective transfer entropy (ETE) and various machine learning algorithms. At first, we explore that the ETE based on 3 and 6 months moving windows can be regarded as the market explanatory variable by analyzing the association between the financial crises and Granger-causal relationships among the stocks. Then, we discover that the prediction performance on the stock price direction can be improved when the ETE driven variable is integrated as a new feature in the logistic regression, multilayer perceptron, random forest, XGBoost, and long short-term memory network. Meanwhile, we suggest utilizing the adjusted accuracy derived from the risk-adjusted return in finance as a prediction performance measure. Lastly, we confirm that the multilayer perceptron and long short-term memory network are more suitable for stock price prediction. This study is the first attempt to predict the stock price direction using ETE, which can be conveniently applied to the practical field.
Highlights
Stock markets have been studied extensively as one of the crucial fields of economy [1]
In this study, we focus on discovering a useful input variable in predicting the stock price direction by utilizing the effective transfer entropy (ETE)-driven network indicator based on five representative machine learning algorithms: logistic regression (LR) as a traditional predictive model, multilayer perceptron (MLP) as a back-propagated neural network, random forest (RF) as a bagging-type ensemble method, XGB as a boosting-type ensemble method, and long short-term memory (LSTM) as a single classifier model
We discover that all five machine learning algorithms have improved the accuracy through the ETE network indicator and suggest that the MLP and LSTM are the most suitable models for predicting future stock price direction predictions when considered the accuracy and adjusted accuracy simultaneously
Summary
Stock markets have been studied extensively as one of the crucial fields of economy [1]. Research has been actively conducted to analyze and predict the stock market based on relationships among the dynamics of stock prices and returns. Since the stocks exhibit diverse interactions, many theoretical or empirical studies of such relationships have provided meaningful implications to investors and policy-makers developing appropriate actions regarding the market condition. The prediction on stock price and the overall market is one of the essential tasks for investors to establish an optimal investment strategy. Many previous studies have utilized concepts in statistical physics such as complex systems and information theory to quantify the correlations among the entities in an. The associate editor coordinating the review of this manuscript and approving it for publication was K.
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
