A carregar...
Publicação
Evaluating credit default risk in P2P lending: a market maturity perspective
| Nome: | Descrição: | Tamanho: | Formato: | |
|---|---|---|---|---|
| 2 MB | Adobe PDF |
Orientador(es)
Resumo(s)
The Peer-to-Peer (P2P) lending industry has grown significantly in recent years, owing its traction to a
growing interest in the market from both individual borrowers and lenders alike. Such platforms have
greatly benefited from the advent of digital transformation and expanding internet footprint across
people of all ages and backgrounds. These platforms leverage novel ways of interacting to facilitate an
alternative means of financing which provides more opportunities to borrowers who may, otherwise
not have access to debt through conventional mechanisms, and more investment opportunities to
lenders who may be seeking to diversify their portfolios.
This study leverages a dataset made available by the Lending Club P2P lending platform, containing
more than 2 million loan entries, amassed over more than 10 years. Being a pioneer in the industry,
Lending Club’s track record and data provide a good window into the inner workings of such platforms
and their ability to assess the creditworthiness of borrowers and loan applications.
We set out to build upon prior work done in this space by understanding and analyzing this dataset to
identify the determinants of default of loans issued through P2P lending platforms. Our analysis is also
employed to create a predictive model which is then tested against our dataset. This approach builds
upon previous studies by outlining an end-to-end process to analyze and assess a platform’s ability to
adequately predict credit default risk.
We have found that, in alignment with prior work, such platforms are indeed able to adequately assess
credit default risk, in the way that grades are assigned to individual loans. The logistic regression model
which we have built has also yielded good results in predicting defaulted loans, while exhibiting
mediocre performance in classifying fully repaid loans as likely cases of default.
Descrição
Dissertation presented as the partial requirement for obtaining a Master's degree in Statistics and Information Management, specialization in Information Analysis and Management
Palavras-chave
Credit Risk Credit Default Default Risk P2P Lending Lending Club Logistic Regression Data Analysis Determinants of Default