Problem 4 (5 pts) Scenario: The following model was built by a banking firm to predict whether a customer is risky to receive a loan or not. The confusion matrices shown represent the model's performance in identifying risky customers in the training data (left) and test data (right). For this problem, you do not have to use R, just evaluate the model presented. (No R script). Variables of Interest risk: 1 if customer is too risky to receive a loan, 0 otherwise ( 0 means a loan can be granted) age: age of customer applicant income: monthly income of customer applicant ficoscore: FICO credit score of customer applicant (high values indicate good credit) For ALL problems, level of significance =0.05 Call: glm(formula = ttrn\$risk ttrn\$age + ttrn\$income + ttrn\$ficoscore, family = "binomial") Deviance Residuals: Coefficients: \begin{tabular}{lrrrr} & Estimate & Std. Error z value & Pr(>z) \\ (Intercept) & 2.700626670 & 0.115592363 & 23.363