Macroeconomic factors, industrial enterprises, and debt default prediction: Based on the VAR-GRU model

Zhenqing Liu, Yi Luo and Mohan Duan

Finance Research Letters, 2025, vol. 78, issue C

Abstract: This study uses a dynamic factor model to construct predictive factors and applies a machine learning-based vector autoregressive model to predict the possibility of corporate bond defaults. The vector autoregressive (VAR) model mainly examines the dynamic interaction relationships among multiple variables, so as to explain the dynamic impacts of various economic shocks on economic variables. It mainly studies the relationships among endogenous variables. Endogenous variables are those variables that are involved in the model and determined within the model system. Exogenous variables, on the other hand, are variables determined by factors outside the model. The Gated Recurrent Unit (GRU), which is a type of Recurrent Neural Network (RNN), can address issues such as the inability of RNNs to have long-term memory and the gradients in backpropagation. It is relatively easy to train. According to data from March 2014 to November 2021, the relevant findings are twofold. 1) A regulatory-based stress test is a crucial tool for measuring the financial sector's resilience in response to challenging macroeconomic conditions. 2) Macroeconomic conditions that may seem unrealistic during economic booms are now often used by regulators as benchmarks for evaluating the losses and capital requirements for market and credit portfolios.

Keywords: Corporate debt default rate; Machine learning; Vector autoregressive model; Dynamic factor model; Integrated feature selection (search for similar items in EconPapers)
Date: 2025
References: Add references at CitEc
Citations:

Downloads: (external link)
http://www.sciencedirect.com/science/article/pii/S154461232500385X
Full text for ScienceDirect subscribers only

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:eee:finlet:v:78:y:2025:i:c:s154461232500385x

DOI: 10.1016/j.frl.2025.107122

Access Statistics for this article

Finance Research Letters is currently edited by R. Gençay

More articles in Finance Research Letters from Elsevier
Bibliographic data for series maintained by Catherine Liu ().