Research on the financial early warning models based on ensemble learning algorithms: Introducing MD&A and stock forum comments textual indicators.

This study analyzes 284 publicly listed companies first designated as ST or *ST between 2015 and 2023. It utilizes two types of textual indicators: Management's Discussion and Analysis (MD&A) and stock forum comments. PCA and MLP are employed for dimensionality reduction. The study compares...

Full description

Saved in:
Bibliographic Details
Main Authors: Zhiheng Zhang, Zhenji Zhu, Yongjun Hua
Format: Article
Language:English
Published: Public Library of Science (PLoS) 2025-01-01
Series:PLoS ONE
Online Access:https://doi.org/10.1371/journal.pone.0323737
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849762407401914368
author Zhiheng Zhang
Zhenji Zhu
Yongjun Hua
author_facet Zhiheng Zhang
Zhenji Zhu
Yongjun Hua
author_sort Zhiheng Zhang
collection DOAJ
description This study analyzes 284 publicly listed companies first designated as ST or *ST between 2015 and 2023. It utilizes two types of textual indicators: Management's Discussion and Analysis (MD&A) and stock forum comments. PCA and MLP are employed for dimensionality reduction. The study compares the recognition performance of single-class models with ensemble learning models while also examining the impact of various base learners and meta-learners on the performance of the ensemble learning model. The findings show that using the two types of textual indicators significantly enhanced the model's accuracy in recognition. The single-class and ensemble learning models demonstrated average improvements of 1.24% and 1.75%, respectively. Notably, stock forum comments outperformed MD&A text. Additionally, the MLP proved more effective in feature processing than PCA. The D-M-BSA-FT model achieved an accuracy of 88.89%. Ensemble learning models outperform single classification models. After introducing textual features, the ensemble learning model achieved an average recognition accuracy of 85.31%, compared to 82.09% for the single classification model. Therefore, the financial warning model developed in this study provides valuable insights for enhancing the accuracy of financial warning identification.
format Article
id doaj-art-5affb2b56d4d40bf9d44eb752d0db0cc
institution DOAJ
issn 1932-6203
language English
publishDate 2025-01-01
publisher Public Library of Science (PLoS)
record_format Article
series PLoS ONE
spelling doaj-art-5affb2b56d4d40bf9d44eb752d0db0cc2025-08-20T03:05:44ZengPublic Library of Science (PLoS)PLoS ONE1932-62032025-01-01205e032373710.1371/journal.pone.0323737Research on the financial early warning models based on ensemble learning algorithms: Introducing MD&A and stock forum comments textual indicators.Zhiheng ZhangZhenji ZhuYongjun HuaThis study analyzes 284 publicly listed companies first designated as ST or *ST between 2015 and 2023. It utilizes two types of textual indicators: Management's Discussion and Analysis (MD&A) and stock forum comments. PCA and MLP are employed for dimensionality reduction. The study compares the recognition performance of single-class models with ensemble learning models while also examining the impact of various base learners and meta-learners on the performance of the ensemble learning model. The findings show that using the two types of textual indicators significantly enhanced the model's accuracy in recognition. The single-class and ensemble learning models demonstrated average improvements of 1.24% and 1.75%, respectively. Notably, stock forum comments outperformed MD&A text. Additionally, the MLP proved more effective in feature processing than PCA. The D-M-BSA-FT model achieved an accuracy of 88.89%. Ensemble learning models outperform single classification models. After introducing textual features, the ensemble learning model achieved an average recognition accuracy of 85.31%, compared to 82.09% for the single classification model. Therefore, the financial warning model developed in this study provides valuable insights for enhancing the accuracy of financial warning identification.https://doi.org/10.1371/journal.pone.0323737
spellingShingle Zhiheng Zhang
Zhenji Zhu
Yongjun Hua
Research on the financial early warning models based on ensemble learning algorithms: Introducing MD&A and stock forum comments textual indicators.
PLoS ONE
title Research on the financial early warning models based on ensemble learning algorithms: Introducing MD&A and stock forum comments textual indicators.
title_full Research on the financial early warning models based on ensemble learning algorithms: Introducing MD&A and stock forum comments textual indicators.
title_fullStr Research on the financial early warning models based on ensemble learning algorithms: Introducing MD&A and stock forum comments textual indicators.
title_full_unstemmed Research on the financial early warning models based on ensemble learning algorithms: Introducing MD&A and stock forum comments textual indicators.
title_short Research on the financial early warning models based on ensemble learning algorithms: Introducing MD&A and stock forum comments textual indicators.
title_sort research on the financial early warning models based on ensemble learning algorithms introducing md amp a and stock forum comments textual indicators
url https://doi.org/10.1371/journal.pone.0323737
work_keys_str_mv AT zhihengzhang researchonthefinancialearlywarningmodelsbasedonensemblelearningalgorithmsintroducingmdampaandstockforumcommentstextualindicators
AT zhenjizhu researchonthefinancialearlywarningmodelsbasedonensemblelearningalgorithmsintroducingmdampaandstockforumcommentstextualindicators
AT yongjunhua researchonthefinancialearlywarningmodelsbasedonensemblelearningalgorithmsintroducingmdampaandstockforumcommentstextualindicators