Open AccessTime‐varying neural network for stock return prediction
Author(s) -
Wong Steven Y. K.,
Chan Jennifer S. K.,
Azizi Lamiae,
Xu Richard Y. D.
Publication year - 2022
Publication title -
intelligent systems in accounting, finance and management
Language(s) - English
Resource type - Journals
eISSN - 1099-1174
pISSN - 1055-615X
DOI - 10.1002/isaf.1507
Subject(s) - stock (firearms) , artificial neural network , financial distress , stock market , computer science , predictive power , econometrics , machine learning , artificial intelligence , context (archaeology) , economics , engineering , mechanical engineering , paleontology , philosophy , epistemology , biology , financial system
We consider the problem of neural network training in a time‐varying context. Machine learning algorithms have excelled in problems that do not change over time. However, problems encountered in financial markets are often time varying . We propose the online early stopping algorithm and show that a neural network trained using this algorithm can track a function changing with unknown dynamics. We compare the proposed algorithm to current approaches on predicting monthly US stock returns and show its superiority. We also show that prominent factors (such as the size and momentum effects) and industry indicators exhibit time‐varying predictive power on stock returns. We find that during market distress, industry indicators experience an increase in importance at the expense of firm level features. This indicates that industries play a role in explaining stock returns during periods of heightened risk.