MACHINE LEARNING ALGORITHM- LOGIT REGRESSION FOR CREDIT RISK ASSESSMENT

Abstract

With the use of Logistic Regression, this study explores the process of credit risk evaluation. The core data for this study comes from ninety individuals who applied for loans at three different private banks in Hyderabad. The rationale of this cram is to uncover foremost drivers of loan default, such as the amount of income, job status, credit history, and debt-to-income ratio at the financial institution. The information is gathered through the use of a structured questionnaire, and then it is analyzed by Logit Regression in order to predict the likelihood of missing payments. The findings offer important insights into the predictive value of financial and demographic characteristics in the progression of appraise credit risk at the individual level. This work makes a contribution to the improvement of datadriven lending choices, which assists financial institutions in optimizing loan approvals while simultaneously reducing the risks of default.