Assignment 8
Victoria Pitonzo & Matthew Volpe
2024-04-11
Objective
- Predict Customer Attrition using the Credit Card data from the AER package, using the classification algorithm from the rpart package.
Data Prep
-Created a new binary variable CreditCardData2 by classifying observations with 0 Customer Attrition as 0 and with one or more Customer Attrition as 1. -Split the data (10127 observations), into training and valdiation datasets (70%/30% , 425/176)
The Tree
Two Interpretations From The Tree
Most Likely Attrited Customer
- Total Transaction Count <55
- Total Revolving Balance Less than 614
- Total Amount Changed for Q1-Q4 Less than 0.65
Unlikely Attrited Customer
- Total Transaction Count <55
- Total Transaction Amount >=5423
Variable Importance
Interesting how 12 months of inactivity would be perceived by many as a predictor for customer attrition. However, the data shows that is is not a good predictor.
Model Accuracy
Using the validation data, it is observed that the model is able to: - Predict correctly 83.9% of Existing Customers and mistake attrited customers as existing customers 15.2% of the time
##
##
## Cell Contents
## |-------------------------|
## | N |
## | Chi-square contribution |
## | N / Row Total |
## | N / Col Total |
## | N / Table Total |
## |-------------------------|
##
##
## Total Observations in Table: 2990
##
##
## | validation_tree$Attrition_Flag
## validation_tree$Attrition_Flag | Attrited Customer | Existing Customer | Row Total |
## -------------------------------|-------------------|-------------------|-------------------|
## Attrited Customer | 454 | 0 | 454 |
## | 2150.935 | 385.065 | |
## | 1.000 | 0.000 | 0.152 |
## | 1.000 | 0.000 | |
## | 0.152 | 0.000 | |
## -------------------------------|-------------------|-------------------|-------------------|
## Existing Customer | 0 | 2536 | 2536 |
## | 385.065 | 68.935 | |
## | 0.000 | 1.000 | 0.848 |
## | 0.000 | 1.000 | |
## | 0.000 | 0.848 | |
## -------------------------------|-------------------|-------------------|-------------------|
## Column Total | 454 | 2536 | 2990 |
## | 0.152 | 0.848 | |
## -------------------------------|-------------------|-------------------|-------------------|
##
##
## Statistics for All Table Factors
##
##
## Pearson's Chi-squared test
## ------------------------------------------------------------
## Chi^2 = 2990 d.f. = 1 p = 0
##
## Pearson's Chi-squared test with Yates' continuity correction
## ------------------------------------------------------------
## Chi^2 = 2982.24 d.f. = 1 p = 0
##
##