In the previous two posts, we explored the data, applied the weight of evidence transformation to our predictors, and eventually built a credit scoring model with logistic regression. Part Three examines other possible methods that can be used to estimate the credit scoring model and demonstrates a credit risk scorecard...
In Part One we demonstrated how to visually and statistically explore the data. In this post, we are going to apply the weight of evidence transformation to our predictors and then use logistic regression to develop our first credit scoring model.
A credit scoring model is a statistical tool widely used by lenders to assess the creditworthiness of their potential and existing customers. The basic idea behind this model is that various demographic attributes and past repayment behaviour of an individual can be utilized to predict hers or his probability of...
blog built using the cayman-theme by Jason Long. LICENSE