Assignment 9
Victoria Apostolico, Jenna Sempkowski
2024-04-11
Predicting Variables and Seperation of Data
- There were 10127 observations in the CreditCardData dataset.
- We will be predicting the customers that leave and the customers that stay.
- We split out data 70/30
- There are 19 predictors.
Variable Importance
Accuracy of Model
##
##
## Cell Contents
## |-------------------------|
## | N |
## | Chi-square contribution |
## | N / Row Total |
## | N / Col Total |
## | N / Table Total |
## |-------------------------|
##
##
## Total Observations in Table: 3110
##
##
## | val$predicted_Attrition_Flag
## val$Attrition_Flag | Attrited Customer | Existing Customer | Row Total |
## -------------------|-------------------|-------------------|-------------------|
## Attrited Customer | 375 | 72 | 447 |
## | 1629.132 | 255.764 | |
## | 0.839 | 0.161 | 0.144 |
## | 0.889 | 0.027 | |
## | 0.121 | 0.023 | |
## -------------------|-------------------|-------------------|-------------------|
## Existing Customer | 47 | 2616 | 2663 |
## | 273.459 | 42.931 | |
## | 0.018 | 0.982 | 0.856 |
## | 0.111 | 0.973 | |
## | 0.015 | 0.841 | |
## -------------------|-------------------|-------------------|-------------------|
## Column Total | 422 | 2688 | 3110 |
## | 0.136 | 0.864 | |
## -------------------|-------------------|-------------------|-------------------|
##
## Financial Impact of Model
I predicted 508 were going to leave
The cost of my program is 508 x 100 = $50,800
We will save 30% of 406 = 121.8, which has a value of 2300 x 121.8 = $280,140
The net of this program is $229,340 from 2990.
Per customer is $75.22 per customer.