Modeling car loan prepayment using supervised machine learning

Abstract

Abstract Logistic regression is a widely used machine-learning model for predicting categorical outcomes. It uses a number of explanatory variables to predict the values of the target variable that can be either binomial or multinomial. It is used in a number of fields such as cancer detection problems, risk modeling and any subject that requires computing the probability of an event occurrence. In this paper, we propose to apply this supervised machine-learning model to study prepayment risk and its determinants concerning car loans based on a number of characteristics.