prediction of default payment next month using a logistic approach

Construction of a robust
prediction model to forecast the
likelihood of a credit card holder
to experience payment defaults in
upcoming months.
Xi global resources
Group of company
March 28, 2024

contents
Introduction
01
Resources
02
Methodology
03
Result and discussion
04
Conclusion
06
Recommendations
05

Overview
 Card payments are essential in digital commerce.
 Card payments integral in digital commerce.
 Show consumer preference and retailer confidence.
 eWallets offer convenient alternative, affirm status.
 Installment options rise, nearly half retailers accept.
 Bank transfers, direct debits less common.
 Longer processing times, lower consumer demand.

Overview cont.'s
 a payment card is one of the best options for
obtaining cash and every year the traditional
cash in the wallet is being displaced more
and more by “plastic money”.
 Number of issued payment cards in 2009–
Q2 2021 (in units) with adjusted lineartrend
and forecast (including 90% confidence
interval)
(2) (PDF) Development of the Payment Cards Market in Poland in the Era of the Covid-19 Pandemic. Available from:
https://www.researchgate.net/publication/361429793_Development_of_the_Payment_Cards_Market_in_Poland_in_the_Era_of_the_Covid-19_Pandemic [accessed Mar 30 2024].

Introduction
Research Motivation and Significant of study
 Build a logistic regression model for classification of customer with a default
payment next month from those without it
 Accurate prediction vital for informed decision-making.
 Payment process affects company finances
significantly.
 Impact on customer relationships and credit
risk.
Problem Statement

Resources
Question addressed
 The only resource provided is the dataset. This dataset contains information on default payments,
demographic factors, credit data, history of payment, and bill statements of credit card clients in
Taiwan from April 2005 to September 2005. The dataset contains 25 variables such as:
 Given historical data and demographic information, can a predictive model effectively estimate
default payment with a high degree of accuracy?
Dataset Information

Resources
Content
There are 25 variables:
1. ID: ID of each client
2. LIMIT_BAL: Amount of given credit in NT dollars (includes individual and family/supplementary credit
3. SEX: Gender (1=male, 2=female)
4. EDUCATION: (1=graduate school, 2=university, 3=high school, 4=others, 5=unknown, 6=unknown)
5. MARRIAGE: Marital status (1=married, 2=single, 3=others)
6. AGE: Age in years
7. PAY_0 to PAY_6 (6 features) Repayment status from September to April, 2005: (-1=pay duly, 1=payment
delay for one month, 2=payment delay for two months, … 8=payment delay for eight months, 9=payment
delay for nine months and above)
8. BILL_AMT1 to BILL_AMT6: Amount of bill statement from April to September, 2005 (NT dollar)
9. PAY_AMT1 to PAY_AMT6: Amount of previous payment from April to September, 2005 (NT dollar)
10. default.payment.next.month: Default payment (1=yes, 0=no)
Dataset Information

Methods
1. Process of assigning labels: Labeling data
2. Preliminary analysis
Exploratory analysis:
Checking the data structure
Detecting missing values
Detecting outliers: boxplot
3. Visualization: A bar chart was used for qualitative variables, while boxplot and density plot was used for the
quantitative or continuous variables.

Methods
 Relationship: Correlation matrix to investigates variable relationships.
 Data partitioning: Dataset split into 75:25 training and test to prevents over fitting.
 Model trained: the model was trained with all features
 Feature selection: Backward selection process was applied to remove insignification feature with a stop
condition set at alpha level of 0.05
 Model evaluation: The model accuracy level was estimated.

Structure of the dataset
Figure 1: Variable contained in the dataset displaying the total number of observation by the
variable type(integer or numeric)

Structure of the dataset
Figure 2: Variable contained in the dataset displaying the percentage of values present(or if
there is any missing values) by total number of observation.

Preliminary Analysis: Exploratory data analysis
Figure 3: Distribution of default payment next month

Figure 3: Distribution of gender by default payment next month

Figure 4: Distribution of marriage by default payment next month

Figure 5: Distribution of education by default payment next month

Figure 6: Distribution of repayment status in September, 2005 by
default payment next month

Figure 7: Distribution of repayment status in August, 2005 by default payment
next month

Figure 8: Distribution of repayment status in July, 2005 by default
payment next month

Figure 9: Distribution of repayment status in June, 2005 by default
payment next month

Figure 10: Distribution of repayment status in May, 2005 by default
payment next month

Figure 11: Distribution of repayment status in April, 2005 by default
payment next month

Figure 12: Distribution of age by default payment next month

Figure 12: Distribution of amount of given credit bill
by default payment next month

Model evaluation
Figure 31:Confusion matrix showing the counts of true positive (TP), true
negative (TN), false positive (FP), and false negative (FN) predictions
made by the model on a dataset
Accura
cy
AUC
Trained
Model
80.84 % 0.73
Retrained
Model
80.91 % 0.72

Recommendation
 Significant Features: The bill amounts (e.g., BILL_AMT1, BILL_AMT3) and previous payment amounts (e.g.,
PAY_AMT1, PAY_AMT2) play a significant role in predicting default offering.
 Payment Status important: It is crucial for the business to closely monitor customers' payment behavior,
especially when there are signs of payment delays or defaults.
 Customer Segmentation: Utilize the insights from the model to segment customers based on their risk profiles.
 Customer Assistance Program: Implement customer assistance programs or financial counseling services to
support customers experiencing financial difficulties.

Conclusion
 In conclusion, the model provide valuable insights into the
factors influencing default payment next month in the dataset.
 The analysis highlights the significance of payment status, age,
bill amounts, and previous payments in predicting default.
 By closely monitoring these factors and adapting strategies
accordingly, businesses can better manage default risks and
improve their financial stability.

prediction of default payment next month using a logistic approach

More Related Content

Similar to prediction of default payment next month using a logistic approach (20)

Recently uploaded (20)

prediction of default payment next month using a logistic approach