Credit Card Default Case for Classification with Random Forests and Boosting. We will use the (Taiwan) Credit
Question:
Credit Card Default Case for Classification with Random Forests and Boosting.
We will use the (Taiwan) Credit card default data with sample size n=30,000 as in your previous case study through read.csv("credit_default.csv"). Conduct a random sampling using your M# to set the seed. Random sample a training data set that contains 80% of the original data, and set the rest 20% aside as a testing data.
Excel info can be found here archive.ics.uci.edu/ml/datasets/default+of+credit+card+clients - > Data Folder - > default of credit card clients.xls
1. Use your 80% training data, build a random forest model on the training data. You may use ">library(randomForest)" and default options from ">randomForest()". What is your in-sample AUC from your random forest model? Please keep four decimals, e.g. 0.7654.
.
2. What is your out-of-sample AUC from your random forest model? Please keep four decimals, e.g. 0.7654.
.
3. Use your 80% training data, build a boosting model on the training data. You may use ">library(adabag)" and default options from ">boosting ()". What is your in-sample AUC from your boosting model? Please keep four decimals, e.g. 0.7654.
.
4. What is your out-of-sample AUC from your boosting model?
Please keep four decimals, e.g. 0.7654.