Data and Code to reproduce results in paper "Network centrality and credit risk: A comprehensive analysis of peer-to-peer lending dynamics"

Data and Code to reproduce results in paper "Network centrality and credit risk: A comprehensive analysis of peer-to-peer lending dynamics"

This repository contains the necessary codes to reproduce results in the paper:

Yiting Liu, Lennart John Baals, Jörg Osterrieder, Branka Hadji-Misheva,
Network centrality and credit risk: A comprehensive analysis of peer-to-peer lending dynamics,
Finance Research Letters,
Volume 63,
2024,
105308,
ISSN 1544-6123,
https://doi.org/10.1016/j.frl.2024.105308.
(https://www.sciencedirect.com/science/article/pii/S1544612324003386)
Abstract: This letter analyzes credit risk assessment in the Peer-to-Peer (P2P) lending domain by leveraging a comprehensive dataset from Bondora, a leading European P2P platform. Through combining traditional credit features with network topological features, namely the degree centrality, we showcase the crucial role of a borrower’s position and connectivity within the P2P network in determining loan default probabilities. Our findings are bolstered by robustness checks using shuffled centrality features, which further underscore the significance of integrating both financial and network attributes in credit risk evaluation. Our results shed new light on credit risk determinants in P2P lending and benefit investors in capturing inherent information from P2P loan networks.
Keywords: Peer-to-Peer lending; Credit-default prediction; Machine Learning; Network centrality

Raw data:

The raw dataset was downloaded on April 22nd, 2022, as a part of Bandora’s daily updated public report.3 Loan starting dates span from June 16th, 2009, to April 21st, 2022. The original dataset covers 231,039 individual borrowers characterized through 112 categorical and continuous variables. Among these loans, 79,424 have been recorded with delayed interest payments according to the platform, while 151,615 loans have no recorded delay on interest payments before the download date of the data. Specifically, the dataset details borrower demographics, financial attributes, and past credit market interactions.

The raw dataset cannot be made public due to the restrictions of the Bondora platform (https://bondora.com/en/terms/): 
13.4 The Portal, Portal's website and the copyright of the contents thereof belong to the Company. The User does not have the right to save, copy, change, transfer, forward or disclose the pages of the Portal for a purpose other than personal use.

Data cleaning:

Bondora.R

further data cleaning.qmd

Thes two files cleans the data, as described in the paper Section 2.2.

Data metadata:

Description cleaned.xlsx

This file describe the meaning of features in the cleaned dataset.

The cleaned dataset, as second-hand data, can also not be made public due to the restriction of the platform.

Modeling:

3rd paper code.qmd

3rd paper code (2).qmd

These two notebooks generate the results presentated in the paper.