The logistic regression model will be built for credit card application scoring model and an ROC curve fit to evaluate the fit of the model.
GC_LR <- glm(good_bad~.,data=GC,family=binomial()).
summary(GC_LR)and identify the significant variables. Also answer the question of whether the model is significant?
LR_Pred <- predict( GC_LR,type='response')
ROCRpackage to set up a prediction object:
GC_pred <- prediction(LR_Pred,GC$good_bad) ...