AI Article Synopsis
- The manuscript presents a new credit default prediction model using TabNeT-Stacking, addressing limitations of traditional models amidst advancing financial technology.
- It employs a custom TabNet structure built with PyTorch and optimizes feature selection and hyperparameters using multi-population genetic and particle swarm algorithms.
- Experimental results indicate that this enhanced model significantly surpasses original credit default prediction models in metrics such as accuracy, precision, recall, F1 score, and AUC.
Article Abstract
With the development of financial technology, the traditional experience-based and single-network credit default prediction model can no longer meet the current needs. This manuscript proposes a credit default prediction model based on TabNeT-Stacking. First, use the PyTorch deep learning framework to construct an improved TabNet structure. The multi-population genetic algorithm is used to optimize the Attention Transformer automatic feature selection module. The particle swarm algorithm is used to optimize the hyperparameter selection and achieve automatic parameter search. Finally, Stacking ensemble learning is used, and the improved TabNet is used to extract features. XGBoost (eXtreme Gradient Boosting), LightGBM (Light Gradient Boosting Machine), CatBoost (Category Boosting), KNN (K-NearestNeighbor), and SVM (Support Vector Machine) are selected as the first-layer base learners, and XGBoost is used as the second-layer meta-learner. The experimental results show that compared with original models, the credit default prediction model proposed in this manuscript outperforms the comparison models in terms of accuracy, precision, recall, F1 score, and AUC (Area Under the Curve) of credit default prediction results.
Download full-text PDF |
Source |
|---|---|
| http://www.ncbi.nlm.nih.gov/pmc/articles/PMC11506879 | PMC |
| http://dx.doi.org/10.3390/e26100861 | DOI Listing |
Publication Analysis
Top Keywords
Similar Publications
African sovereign risk premia and international market assets: A relook under the COVID-19 outbreak.
Heliyon
November 2024
Department of Accounting and Finance, School of Business, Ghana Institute of Management and Public Administration (GIMPA), Achimota, Ghana.
Using the wavelet multiscale coherence technique, the paper examines the interdependences between global market assets, sovereign credit default swap (CDS) and yield-to-maturity on bond spread for African economies from January 2019 to March 2023. The empirical results primarily reveal a high level of coherence between global market assets and Africa's sovereign CDS and yield-to-maturity, although varying across countries and time-frequency domain. Notably, the high degree of coherence and the positive co-movement between sovereign CDS and global market assets are more concentrated at the medium and upper frequency bands.
View Article and Find Full Text PDFAdvancing financial resilience: A systematic review of default prediction models and future directions in credit risk management.
Heliyon
November 2024
Department of Business Administration, IQRA University Karachi, Pakistan.
This research presents a systematic review of a substantial body of high-quality research articles on Default Prediction Models published from 2015 to 2024. It is a comprehensive analysis of a DPM wide spectrum approaches including Textual Models, Systematic Review Studies, Hybrid Models, Intelligent Models and Statistical Models. The reason behind this study is rooted in the critical need to mitigate and understand the credit default risk that poses a significant threat to financial stability worldwide.
View Article and Find Full Text PDFThis paper, starting from the effects of financial regulatory policies, considers the interaction between Chengxin_Moody and Lianhe_Fitch with the dual rating system and the multi-rating system, constructs a new ordered Logit model, and attempts to explore the impact of the Notice, the dual rating system and the multiple rating system on the probability of Chinese corporate bond defaults, rating upgrades, rating downgrades, and the magnitude of credit rating migrations. This study compares the effectiveness of different rating regulatory systems. Using nine thousand two hundred and sixty-two data of Chinese corporate bonds as the research samples.
View Article and Find Full Text PDFDo typhoon storm surges affect bank performance in coastal regions? Empirical evidence from China.
Mar Pollut Bull
December 2024
School of Economics, Ocean University of China, Qingdao 266100, Shandong, China. Electronic address:
Typhoon storm surges cause significant damage to economic activities and have a huge impact on the financial sector. Using panel data of 170 local commercial banks in China's coastal regions from 2008 to 2022, this study explores the impact of typhoon storm surges on bank performance. The study finds that typhoon storm surges significantly reduce bank performance, with a greater impact on banks in the southern and eastern marine economic circles, as well as those primarily serving rural and "Tam Nong" economies.
View Article and Find Full Text PDFA dynamic receptive field and improved feature fusion approach for federated learning in financial credit risk assessment.
Sci Rep
November 2024
School of Information Engineering, Hubei University of Economics, Wuhan, 430205, China.
Federated Learning (FL) uses local data to perform distributed training on clients and combines resulting models on a public server to mitigate privacy exposure by avoiding data sharing. However, further research indicates that communication overheads continue to be the primary limitation for FL relative to alternative considerations. This is especially true when training models on non-independent and identically distributed data, such as financial default risk data, where FL's computational costs increase significantly.
View Article and Find Full Text PDFWant AI Summaries of new PubMed Abstracts delivered to your In-box?
Enter search terms and have AI summaries delivered each week - change queries or unsubscribe any time!