Predicting stock returns using machine learning combined with data envelopment analysis and automatic feature engineering: A case study on the Vietnamese stock market

Hoang Thanh Nhon, Nga Do-Thi and Thao Nguyen-Trang

PLOS ONE, 2025, vol. 20, issue 9, 1-16

Abstract: In financial markets, predicting stock returns is an essential task for investors. This paper is one of the first studies using business efficiency scores calculated from data envelopment analysis to predict stock returns. In the meantime, this is also one of the first studies to comprehensively investigate the performance of machine learning models and automatic feature engineering techniques in the context of predicting returns in the Vietnamese stock market. Specifically, the data from 2019 to 2024 of 26 real-estate enterprises on Ho Chi Minh Stock Exchange are collected. Based on relevant technical indicators, fundamental indicators, and business efficiency scores, a comparison of various machine learning models’ performance is provided. The results indicate that incorporating business efficiency scores significantly enhances the models’ accuracy. For example, the deep neural network model shows a decrease in RMSE from 0.926 to 0.375, MAE from 0.337 to 0.196, and MAPE from 134.63 to 114.71. Furthermore, the gradient boosted tree model, when combined with business efficiency scores and automatic feature engineering, achieves the best results, yielding an MAE of 0.122 and an MAPE of 103.19. The obtained results reveal a significant improvement in terms of accuracy when using the business efficiency score with the automated feature engineering technique.

Date: 2025
References: Add references at CitEc
Citations:

Downloads: (external link)
https://journals.plos.org/plosone/article?id=10.1371/journal.pone.0332154 (text/html)
https://journals.plos.org/plosone/article/file?id= ... 32154&type=printable (application/pdf)

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:plo:pone00:0332154

DOI: 10.1371/journal.pone.0332154

Access Statistics for this article

More articles in PLOS ONE from Public Library of Science
Bibliographic data for series maintained by plosone ().