Using Alternative Data to Evaluate Credit Risk
Challenge
Currently, female entrepreneurs are more likely to get lower premiums, higher interest rates, and increased penalties for mistakes, due to out-of-date, gender-biased lending technology and practices. One billion women remain outside the formal financial system today. A solution for this is particularly important in the present climate as emerging markets look to recover from the effects of COVID-19.
To address these issues, Women’s World Banking, in partnership with the University of Zurich, is exploring the implications of AI-based modeling and credit scoring on women’s financial inclusion.
As part of this project, we will work with a financial services provider to develop a gender-aware algorithm for credit scoring. Our data scientists are looking for help from a credit risk analyst to help audit the usability of the scoring model for the financial services provider.
Expected outcome
Weekly feedback on credit scoring model.
Resources provided
Customer data, scoring models, and code.
Program structure
Évaluer le modèle de notation de crédit
Appliquer les meilleures pratiques en matière de notation de crédit au modèle de notation de crédit
Estimer les répercussions commerciales du modèle de notation de crédit
Rédiger chaque semaine des comptes rendus d'une page avec évaluation.
Compétences
Les langues
Nonprofit mission
Our mission is to expand the economic assets, participation, and power of low-income women and their households by helping them access financial services, knowledge, and markets.
Publié par
