Incorporating multilevel macroeconomic variables into credit scoring for online consumer lending

作者：

Highlights：

• Multilevel macroeconomic variables (MVs) are first applied to credit scoring.

• We emphasize the credit risk of online consumer lending in developing economies.

• A structural model is developed to analyze the determinants of consumer credit risk.

• A Bayesian MV selection and lag optimization approach is proposed.

• The incorporation of multilevel MVs outperforms the benchmarks significantly.

摘要

•Multilevel macroeconomic variables (MVs) are first applied to credit scoring.•We emphasize the credit risk of online consumer lending in developing economies.•A structural model is developed to analyze the determinants of consumer credit risk.•A Bayesian MV selection and lag optimization approach is proposed.•The incorporation of multilevel MVs outperforms the benchmarks significantly.

论文关键词：Credit scoring,Online consumer lending,Consumer credit risk,Macroeconomic variables,FinTech

论文评审过程：Received 15 March 2021, Revised 30 May 2021, Accepted 20 September 2021, Available online 22 September 2021, Version of Record 7 October 2021.

论文官网地址：https://doi.org/10.1016/j.elerap.2021.101095