Prof Kanshukan Rajaratnam (School for Data Science and Computational Thinking, Stellenbosch University)
Consumer Lending Models
When faced with a pool of loan applications, bank managers are required to accept or reject each application. In order to help make these decisions, scorecards are used to forecast the risk of each applicant and hence, a probability of whether the loan would be paid back or not. Historical data of past and existing account holders are used to build these risk scorecards and then, new applicants are scored to forecast their risk. These scorecards help to reduce the asymmetry of information available to the lender and to the applicant. Past research has shown improvements in scorecards will require new variables/data rather than new modelling methods. In this exploratory research, we devise a game that reveals the decision heuristics of applicants and then discuss whether this new data (decision heuristics) could be used as an input into scoring models.
Kanshukan Rajaratnam is the Director of the School for Data Science and Computational Thinking at Stellenbosch University. He holds degrees in chemical engineering (BScEng - UKZN), Industrial Engineering (MSc) and Systems Engineering (PhD). After having spent many years studying various engineering discplines, he joined the Department of Finance and Tax at the University of Cape Town (from 2011 to 2019). He is a member of the South African Young Academy of Science (SAYAS) and has served two years on the exco of SAYAS.