1.
Introduction
The credit default and economic data sets are the two under investigation. Financial institutions like
banks and credit card firms may use the data to assess a person's risk of default when they ask for credit
lines or loans. This Economists can use this method to study trends in wage growth. Decision trees are a
form of evaluation of these two problem sets.
2.
Data Preparation
The set of questions you receive requires you to examine some key variables. Determine and
elucidate these variables. In your analysis, answer the following questions: This data set's key
variables are default, credit utilization, missed payments, assets, marriage, education, sex, and
age. In this data set, there are eight columns and six hundred rows.
3.
Classification Decision Tree
The initial data set contains 600 rows. 180 are the validation data sets, while 420 are the training data
sets. The table below shows the variable's classification decision tree based on the provided predictors.
CP nsplit rel error xerror xstd 2 0.048913 1 0.206522 0.20652 0.031951 3 0.016304 3 0.108696 0.10870
0.023719 4 0.010000 4 0.092391 0.11413 0.024275 In the pruned tree, the cp value is 0.793478. The
resulting decision tree has an appropriate cp value of 0.793478; the plot is shown below.
Reporting Results
Evaluating Utility of Model
Evaluate the utility of the classification decision tree. Address the following questions in your analysis:
The Obtained report for the true positives, negatives and false positive and negatives are as follows:
true positives, = 84
true negatives, = 87
false positives = 5
false negatives = 4
Report the following:
○ Accuracy = 19 20 = 0.95 ○ Precision = 84 89 = 0.94 ○ Recollect = 21 22 = 0.95
2
