Prediction of Installment Default Risk of Financial Products Based on Consumers’ Buy-Now-Pay-Later Behavioral Characteristics

DOI

10.2991/978-94-6239-801-6_128How to use a DOI?

Keywords

BNPL; Consumer behavior; Credit risk prediction; K-means clustering; Logistic regression

Abstract

With the widespread adoption of buy-now-pay-later (BNPL) in the consumption of high-priced goods, installment payments have not only enhanced the accessibility of consumption but also increased the exposure to default risks. Moreover, a single traditional indicator is no longer sufficient to accurately depict the heterogeneity of risks. This article focuses on the installment payment scenarios for high-priced goods, using 7,380 consumer transaction data as a sample. By combining K-means clustering and Logistic regression, it conducts a study on default risk prediction from the perspective of behavior. The research classifies consumers into three categories based on their consumption behavior characteristics. These are: the high-income, high-brand preference, high-consumption group (Cluster 1), the low-income, low-brand preference, low-consumption group (Cluster 0), and the group with balanced characteristics in the middle (Cluster 2). The results show that there are significant differences in risk among the three groups. The loss rate of Cluster 1 is the highest (0.22), followed by Cluster 0 (0.17), and Cluster 2 has the lowest rate (0.10). High-spending customer segments do not necessarily represent the lowest-risk categories. This study reveals the correlation between consumption behavior patterns and installment default risk, thereby refining the risk identification framework for installment financing. It provides empirical support for financial institutions to implement tiered risk management and differentiated risk control strategies.

Copyright

© 2026 The Author(s)

Open Access

Open Access This chapter is licensed under the terms of the Creative Commons Attribution-NonCommercial 4.0 International License (http://creativecommons.org/licenses/by-nc/4.0/), which permits any noncommercial use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license and indicate if changes were made.

Download article (PDF)

TY  - CONF
AU  - Ruiyang Hong
PY  - 2026
DA  - 2026/10/23
TI  - Prediction of Installment Default Risk of Financial Products Based on Consumers’ Buy-Now-Pay-Later Behavioral Characteristics
BT  - Proceedings of the 2026 2nd International Conference on Financial Innovation and Marketing Management (FIMM 2026)
PB  - Atlantis Press
SP  - 1236
EP  - 1246
SN  - 2352-5428
UR  - https://doi.org/10.2991/978-94-6239-801-6_128
DO  - 10.2991/978-94-6239-801-6_128
ID  - Hong2026
ER  -