2. 1st Case Study : Credit Rating Model
Borrowers and Factoring (Accounts Receivable Financing) pages 3 – 10
2nd Case Study : Credit Scoring Model
Automobile Leasing pages 11 – 20
3rd Case Study : The Validation of Internal Rating Systems pages 21 – 28
Credit Risk Model : What is Credit Risk Model? page 29
What Properties to be expected! page 29
Application page 30
AGENDA
3. 1st Case Study
Credit Rating Model
Borrowers and Factoring
(Accounts Receivable Financing)
3
4. Credit Rating Model ….. (1)
Model Development
Portfolio
Analysis
Performance
Test
Data
Analysis
Data
Preparation
Implementation
Recommendation
Calibration and
Mapping
Data
Cleansing
Data
Partition
Univariate Correlation Multivariate
Accuracy Blind Test
4
5. Credit Rating Model ….. (2)
Portfolio Analysis
Distribution of Financial Statements and Default Data
-Year; Business Type / Group / Concentration; Size
Performance
Window
Bad
Definition
Coverage
Number of
Accounts
5
6. Credit Rating Model ….. (3)
Data Preparation
Data Cleansing Data Partition
Sample
Grouping
Dimension of
Sampling
Existing and
Completeness of
Data
Variables
-FinancialRatio
-Qualitative Data
Correctionof
Data
Logic
Loan Status
comparedto
FinancialPerformance
In Sample /
Out-of-Sample
Out-of-Universe
Out-of-Time
Product Type
Business Type
Asset Size
6
7. Credit Rating Model ….. (4)
Data Analysis
Univariate Correlation Multivariate
Logistic
Regression
Screening
Criteria
Multiple Logistic
Regression
Quantitative
Qualitative
Make
Business Sense
Powerful
Understandable /
Intuitive
Enough
observation to
develop and
validate model
7
12. Credit Scoring Model
Scorecard Development
Portfolio Analysis Model Development
Portfolio
Distribution
Vintage
Analysis
Test on Similarity and/or Differentiation
of Bad Rates
Data
Gathering
Model
Building
Data
Cleansing
Variable
Classing
Preliminary
Variable Selection
Purposes of analyzing portfolio:
-To understand the overall picture of portfolio
-To know the portfolio’s default rates in terms of Marginal and Cumulative Default Rates
-To help set the sample group to be collected for model development
12
13. Model Development ….. (1)
Data Gathering
Performance
Window
VariablesSample
Size
Bad
Definition
Dependent
Variables
Independent
Variables
Borrower’s
Characteristics
Loan
Status
Facility’s
Characteristics
Collateral’s
Characteristics
13
14. Model Development ….. (2)
Data Cleansing
Data Exploration Problem and Solution
Number of Fields Number of
Complete Cases
Missing
Value
Error from Data
Transformation
Extreme
Value
Vague /
Unclear Data
Reliability of
Data
Missing
Data
Discrepancy of
Data
14
15. Model Development ….. (3)
Variable Classing
Manage the
number of
attributes per
characteristics
Make the Weight
of Evidence – and
thereby the
number of points
in the scorecard –
vary smoothly or
even linearly
across the
attributes
Select predictive
characteristics
Improve the
predictive power
of the
characteristics
Classing is process of automatically and / or interactively binning and grouping interval, nominal, or
ordinal input variables in order to
15
16. Model Development ….. (4)
Variable Selection
Preliminary Statistical
Missing
Value
Consistency
Completeness and relationship with
other variables
Univariate
data analysis
Multivariate
data analysis
16
17. Model Development ….. (5)
Model Building
Why Logistic Regression!
-It can handle discrete variable or qualitative variable.
-The dependent variable need not to be normally distributed.
-The dependent variable need not to be homoscedastic for each level of the independents.
-Normally distributed error terms are not assumed.
-It does not require that the independents be interval.
17
18. Model Development ….. (6)
Model Building
Sample Accuracy
Training
Sample
Out-of-time
Sample
Type I & II
errors
K-S
Statistics
Testing
Sample Discriminatory Power
of the model : The
maximum difference
between the
cumulative percent
good distribution and
the cumulative
percent bad
distribution
18
19. Model Development ….. (7)
Criteria for Model Selection
Accuracy : Type I & II
errors; K-S statistics
Including or excluding
Policy Indicators in
the model
Number of Variables
in the model
Consistency of testing results
among Training, Testing and
Out-of-time Samples
ExampleofGainTable
19
22. Scope of Work and Validation Aspects
Scope of Work
Analyze the discriminatory power
of rating models
ValidationAspects
Analyze the stability
of rating models
Analyze the connection between
PD and Grade
Analyze the models design
Analyze the rating process
22
23. Validation Method
Study and Analyze Data
from Documents
Interview and
Site VisitDefault Probability Model
Validation
Statistical and
Mathematical
Tests
23
24. Validation Results
Quantitative : Discriminatory Power
Type I & II errors in theory
RelativeFrequency
Rating Class
Density Function for Bad Cases
Density Function for Good Cases
Cut-Off Point
Type I error Type II error
Project Financing <15M;
Hire Purchase; Leasing
Project Financing <15M
Hire Purchase and Leasing
Project Financing >= 15M Factoring
Type I error of each model
24
27. Validation Results
Quantitative : Calibration
Actual Default Rate (%) of each model compared to
Implied PD (%) by CQC Grade
PF < 15M
PF >= 15M
HP: LS
PF < 15M;
HP; Leasing
Implied PD
Actual Default Rate (%) of each model compared to
S&P’s Default Rate by Rating
PF >= 15M
HP: LS
S&P’s
Default Rate
PF < 15M
PF < 15M;
HP; Leasing
27
29. Credit Risk Model ….. (1)
What is Credit Risk Model?
A tool used to evaluate the level of risk associated with applicants or borrowers.
It consists of a group of characteristics, statistically determined to be predictive in separating “good”
and “bad” accounts.
It provides statistically odds or probability that an applicant or borrower with any given rating or score
will be “good” or “bad”.
What Properties to be expected!
Understandable
Powerful
Calibrated
Empirically validated
29
30. Credit Risk Model ….. (2)
Application
Origination Decisions
Given the risk and a fixed price, is the asset worth taking?
Given the risk, what price is required to make the asset worth buying?
Portfolio Optimization
To reduce the portfolio’s risk, concentrations of risk and how the risk can be diversified must be known.
Capital Management
To set capital, the loss level is needed.
Credit Process Management
To gain the efficiencies of application processing that comes through automation.
To gain control and consistency in lending practices for the entire credit portfolio.
To identify the variables which are important in the credit evaluation process
To improve delinquency statistics while maintaining desired approval rates
30
31. Credit Risk Model ….. (3)
Credit Risk Model
only classifies and predicts risk;
It does not tell the lender
how to manage it.
31