The Value of Big Data for Credit Scoring: Enhancing Financial Inclusion using Mobile Phone Data and Social Network Analytics (2002.09931v1)

Abstract: Credit scoring is without a doubt one of the oldest applications of analytics. In recent years, a multitude of sophisticated classification techniques have been developed to improve the statistical performance of credit scoring models. Instead of focusing on the techniques themselves, this paper leverages alternative data sources to enhance both statistical and economic model performance. The study demonstrates how including call networks, in the context of positive credit information, as a new Big Data source has added value in terms of profit by applying a profit measure and profit-based feature selection. A unique combination of datasets, including call-detail records, credit and debit account information of customers is used to create scorecards for credit card applicants. Call-detail records are used to build call networks and advanced social network analytics techniques are applied to propagate influence from prior defaulters throughout the network to produce influence scores. The results show that combining call-detail records with traditional data in credit scoring models significantly increases their performance when measured in AUC. In terms of profit, the best model is the one built with only calling behavior features. In addition, the calling behavior features are the most predictive in other models, both in terms of statistical and economic performance. The results have an impact in terms of ethical use of call-detail records, regulatory implications, financial inclusion, as well as data sharing and privacy.

• The paper demonstrates that integrating mobile call records into credit scoring models significantly improves AUC and profit margins.
• It employs call-detail records and social network analytics to derive influence scores that serve as powerful predictors of default behavior.
• The study reveals economic gains in risk evaluation and highlights the potential for financial inclusion while addressing ethical data concerns.

The Value of Big Data for Credit Scoring

The paper "The Value of Big Data for Credit Scoring: Enhancing Financial Inclusion using Mobile Phone Data and Social Network Analytics" delivers a comprehensive paper that highlights the potential of leveraging innovative data sources in credit scoring. Unlike many traditional studies that focus on refining analytical techniques, this work pivots towards integrating alternative data sources, namely mobile phone call-detail records (CDR), into credit evaluation processes.

Research Objectives and Methodology

The paper sets out to explore three main research questions:

1. The added value, both in terms of AUC and profit, of incorporating call data into credit scoring models.
2. Whether call data can serve as a substitute for the conventional data traditionally used in credit scoring.
3. The mechanisms underpinning default behavior propagation within a call network.

The researchers employ a dataset that intricately combines mobile phone call records with conventional banking information, such as sociodemographic data and transaction histories. By constructing call networks utilizing CDRs, the paper applies advanced social network analytics to propagate influence from known defaulters across the network, generating influence scores that serve to enhance the predictive power of traditional scorecards. Statistical measures such as the area under the receiver operating curve (AUC) and new measures like the Expected Maximum Profit (EMP) are used to assess the performance of these enriched models.

Key Findings and Statistical Insights

The inclusion of mobile phone data in credit scoring models yielded significant improvements in the statistical performance, as noted by the increased AUC values. Models incorporating calling behavior as features were shown to be the most effective on several accounts. Notably, these variables contributed substantially to the improvement in credit scoring models compared to traditional sociodemographic and financial data. Furthermore, variables derived from calling behavior, specifically those capturing the social network influence, emerged as pivotal predictors.

Economic Implications and Profitability

From an economic perspective, the analysis conducted through the Expected Maximum Profit (EMP) measure demonstrated substantial potential for enhanced profitability when call data was included. Interestingly, models that exclusively leveraged calling behavior features recorded higher profit margins than those using traditional data, underscoring the economic viability of incorporating mobile data into credit assessment processes.

Broader Implications and Future Directions

The research has significant implications for both practitioners and policymakers. For financial institutions, the ability to utilize alternative data sources such as CDR presents an opportunity to refine risk assessment strategies, potentially broadening access to credit—especially to individuals without established credit histories. This could be transformative in emerging markets where traditional credit data is limited.

However, the paper does not shy away from addressing potential ethical and regulatory concerns. The use of personal network data necessitates careful consideration of privacy regulations, necessitating frameworks that ensure the ethical deployment of such models.

Looking ahead, future research could delve into the implications of using different propagation methods within network analyses or extending these methodologies to diverse loan products beyond credit cards. Furthermore, gathering comparative data from multiple regions could enhance the external validity of these findings, offering deeper insights into their applicability across different financial landscapes.

In conclusion, this paper convincingly argues for the inclusion of alternative data sources in credit scoring processes, not only as a means to boost statistical and economic performance but also as a pathway to promote greater financial inclusion.