Loan officer at any bank would be interested in detecting the factors which can identify people who are likely to default on loans, consequently good and bad credit risks. Moreover, he will also be interested in designing model which can predict chances of default with reasonable accuracy. A crucial step in the loan application process is financial risk assessment, which helps financial institutions minimize credit risk and allocate resources as efficiently as possible. This paper creates a predictive framework for estimating the probability of loan defaults by combining factor analysis and logistic regression. While Factor Analysis lowers dimensionality and finds latent characteristics influencing financial risk, like credit history and debt-to-income ratio. Logistic Regression is a strong classification technique is used to estimate default probabilities by balancing sensitivity and specificity, the suggested model showed excellent predictive accuracy. It offers a data-driven and interpretable way to expedite loan approval decisions, guaranteeing a more fair and effective procedure.
Kushal Dutia (Fri,) studied this question.