Construction and optimization of financial risk management model based on financial data and text data influencing information system

• Main Authors: Hui Huang, Thien Sang Lim
• Format: Article
• Language: English
• Published: IADITI 2024
• Subjects: HG1-9999 Finance; P302-302.87 Discourse analysis
• Online Access: https://eprints.ums.edu.my/id/eprint/42879/1/FULL%20TEXT.pdf; https://eprints.ums.edu.my/id/eprint/42879/; https://doi.org/10.55267/iadt.07.14767

A-share companies must manage financial risk to succeed. Textual data insights can great lyimpact risk assessment results, although most risk management systems focus on quantitative financial assessments. This research constructs and enhances information system financial risk management models employing financial and textual data, including MD&A narratives, to fill this gap. Westudy how textual data aids financial risk management algorithms' risk prediction. Textual andfinancial research on 2001–2022 Shenzhen and Shanghai Stock Exchange companies is used. This study found financial and non-financial data models more predictive. Qualitative textual informationis used in financial risk assessment to improve risk prediction algorithms. MD&A texts, sentiment analysis, andreadability signal risk. Internet forum discussions are linked to financial risk, but media coverageisnot. These unconventional data sources evaluate financial risk. The research shows that A-sharecorporations manage financial risk. The study advises merging qualitative textual data withfinancialmetrics to solve literature gaps and improve risk management. Shenzhen and Shanghai StockExchange statistics suggest MD&A storylines might strengthen financial risk management models.Study shows readability and sentiment analysis increase risk model prediction. The study foundthat textual material affects financial risk, therefore risk assessment should include non-financial information. This complete risk management technique may assist A-share listed companies navigate financial markets and make smarter decisions using quantitative financial data and qualitative textual insights. This study implies textual data may help financial risk algorithms. MD&As helpcompanies identify and manage financial risk. More study is needed to discover new textual elements and strengthen context-specific risk management frameworks.