Title:	METHOD OF ESTIMATING PRESUMED INCOME FOR NATURAL PERSONS IN THE OPEN BANKING ENVIRONMENT
Author(s):	Flávio Henrique de Souza Gonçalves, João Carlos Félix Souza, Diogo Suzart Uzeda Picco and André Nunes Maranhão
ISBN:	978-989-8704-36-8
Editors:	Piet Kommers, Tomayess Issa, Adriana Backx Noronha Viana, Theodora Issa and Pedro Isaías
Year:	2021
Edition:	Single
Keywords:	Credit Risk Management, Credit Scoring, Presumed Income, Quantile Regression, Analytical Intelligence, Open Banking
Type:	Full
First Page:	84
Last Page:	90
Language:	English
Cover:
Full Contents:	click to dowload
Paper Abstract:	One of the fundamental pieces of information in the credit granting process of a financial institution is the individual's ability to honor commitments assumed. In this context, their individual income is of utmost importance to properly determine one's payment capacity and the volume of resources to grant to each client. A statistical model capable of estimating individual's income it's of main relevance beyond the credit process, it comprises regulatory requirement, customer loyalty/prospecting, money laundering combating, validation of information provided without proper proof and subjected to operational risk at the time of data internalization, among others. The possession of reliable and up-to-date income information becomes, therefore, a relevant competitive advantage for financial institutions. With the beginning of Open Banking in Brazil, the challenge of estimating presumed income of natural persons, using a proprietary methodology, could be the differential for financial institutions and fintechs in this new arena of the National Financial System (SFN). In view of this new ecosystem for sharing customer data, the need to develop statistical models was identified to improve the consistency in recording information, mitigating credit and operational risks, prospecting new clients and increasing operational efficiency by preventing fraud. The present work aims at describing the steps for the elaboration of a predictive model of Presumed Income for Individuals. It uses analytical intelligence techniques, particularly quantile regression, which is applied to a client database of a large Brazilian financial institution, testing its use in a credit score model in policies for granting credit to new clients in the Open Banking environment.
	Go Back