BIMSA >
BIMSA Digital Economy Lab Seminar
High-dimensional imbalanced financial distress prediction: A multi-heterogenous self-paced ensemble learning framework
High-dimensional imbalanced financial distress prediction: A multi-heterogenous self-paced ensemble learning framework
Organizers
Speaker
Time
Monday, September 30, 2024 3:20 PM - 4:20 PM
Venue
A3-2-303
Online
Zoom 230 432 7880
(BIMSA)
Abstract
Financial distress prediction (FDP) is a crucial research field for researchers, stakeholders, and government regulators. However, researchers often face challenges such as high dimensional problem, class imbalanced problem, and parameter optimization problem when dealing with FDP tasks. This makes it difficult for the model to effectively identify high-risk companies. To address these problems, we propose a novel multi-heterogenous self-paced ensemble learning framework for financial distress prediction, named FinMHSPE. Our proposed model employs a pair comparison of multiple time span data and the maximum relevance and minimum redundancy (mRMR) method to determine the optimal feature subset, which can effectively overcome the high dimensional problem. We also propose the MHSPE model to iteratively learn the most informative majority data samples for dealing with the class imbalanced problem. Finally, we employ the particle swarm optimization (PSO) algorithm to determine the optimal parameters of the ensemble learning method. To verify the effectiveness of our proposed model, we conduct a series of experiments on the financial dataset of Chinese listed companies. The empirical results demonstrate that our model significantly outperforms competing models. Furthermore, we identify some useful features for improving FDP model performance, such as the features of the company’s information and growth capability.
Speaker Intro
Ruize Gao is Post-Doctoral Fellow at Beijing Institute of Mathematical Sciences and Applications and Tsinghua University. His research interests include digital economy, data economy and data mining. He has published in leading journals such as Information Sciences, Knowledge-based Systems, Expert Systems with Applications, Technology in Society, International Journal of Accounting Information Systems. He hosts one funding project under the China Postdoctoral Science Foundation.