Retail Credit Risk

Question 1

define retail lending

Answer 1

exposure to an individual/small business, and guaranteed by such person

Question 2

what are 4 examples of retail lending.

Answer 2

credit cards
residential mortgages
small business facilities
installment loans

Question 3

what are two characteristics of retail lending

Answer 3

low individual exposure
managed collectively rather than individually

Question 4

what is a credit risk score

Answer 4

a total number of points that predicts a borrower’s future repayment performance based on historical information

Question 5

what is a scorecard

Answer 5

a mathematical algorithm used to generate a score for rank-order risk analysis

Question 6

what are scorecards used for

Answer 6

1. lending decisions
2. mitigation of portfolio credit risk

Question 7

what are two benefits of using a scorecard

Answer 7

easy to interpret
easy to monitor

Question 8

what are the 6 stages in model development

Answer 8

1. business objectives
2. data preparation
3. model development
4. model approval
5. model deployment
6. monitoring

Question 9

what are 3 aspects of business objectives

Answer 9

1. key issues
2. expectations for the model
3. structure

Question 10

define key issues

Answer 10

trends, challenges and concerns outlined by the business

Question 11

define structure

Answer 11

project team members, data and timeline

Question 12

what are the 5 C’s of data preparation

Answer 12

1. Comprehensiveness
2. Clean
3. Consistent
4. Current
5. Caretaking

Question 13

define comprehensiveness

Answer 13

ensuring the data captures the full scope and complexity of the underlying information

Question 14

define clean

Answer 14

ensuring the accuracy of the data

Question 15

define consistent

Answer 15

ensuring the uniformity of the data across different sources.

Question 16

define current

Answer 16

ensuring the data is up to date

Question 17

define caretaking

Answer 17

the ongoing management of the data to preserve its quality

Question 18

what are 6 aspects of the data preparation in the model development lifecycle

Answer 18

the 5 C’s
exclusion criteria
timeframe
defining the target and explanatory variables
segmentation (# of models)
sampling

Question 19

what are three sources of exclusion criteria

Answer 19

scope
data errors
operational

Question 20

what two periods are involved in the timeframe of model creation

Answer 20

1. observation period
2. performance period

Question 21

what are two aspects of the observation periods

Answer 21

1. for explanatory variables
2. should be representative of the current/future environment

Question 22

what are two aspects of performance periods

Answer 22

1. for the target variable
2. should be long enough to have a sufficient number of defaults.

Question 23

what are the two modeling techniques

Answer 23

1. industry standard
2. other methodologies

Question 24

compare the advantages of the two modeling techniques

Answer 24

industry-standard:
1. few variables
2. expert judgment

other methods:
1. many variables
2. one step for variable reduction and model fitting
3. adaptive learning

Question 25

compare the disadvantages of the two modeling techniques

Answer 25

industry-standard: 1. few variables 2. distributional assumptions 3. separate steps for variable reduction and model fitting other methods: 1. many variables 2. risk of overfitting

Question 26

what are the 5 steps of the industry standard model development technique?

Answer 26

1. variable transformation 2. variable reduction 3. model fitting 4. scorecard scaling 5. scorecard assessment

Question 27

what technique can be used in variable transofrmation

Answer 27

weight of evidence

Question 28

define variable reduction

Answer 28

removing any variable that cannot be used or doesnt make sense

Question 29

what are two techniques for variable reduction

Answer 29

1. grouping 2. variable clustering

Question 30

what is grouping

Answer 30

creating bins within a variable

Question 31

what are three benefits of grouping?

Answer 31

i. Accounts for non-linear relationship between the target and explanatory variables. ii. Accounts for outliers iii. Allows for the treatment of missing values as a separate category.

Question 32

how should grouping be performed?

Answer 32

1. Start by creating 20 equal bins. 2. Calculate the WOE of each bin. 3. Collapse bins with similar WOE. 4. Remove variables with weak IV.

Question 33

what is variable clustering

Answer 33

grouping correlated variables together such that variables within a cluster are highly correlated and variables between of clusters are uncorrelated two reduce the multicollinearity of the model.

Question 34

which two variables should represent the cluster then using variable clustering?

Answer 34

1. the variable with the highest IV 2. the variable with the lowest 1-R^2

Question 35

what are two aspects of model fitting in the industry standard technique?

Answer 35

variable selection: forward, backwards, ridge lasso assumptions that historical experiences predict future behaviour and that consumer behaviour will not change significantly

Question 36

define scorecard scaling when using the industry standard technique

Answer 36

raw scores are scaled to a three digit number

Question 37

what is the formula in score in scorecard scaling

Answer 37

score=offset+(factor⋅ln⁡(2⋅odds) )-PDO

Question 38

what are the 3 types of scorecard assessment

Answer 38

1. rank ordering 2. population stability 3. benchmarking

Question 39

what are the 5 evaluation metrics used in rank ordering scorecard assessment

Answer 39

1. KS statistic 2. misclassification 3. ROC curve 4. accuracy ratio 5. lift chart

Question 40

what does population stability do

Answer 40

quantify population differences by measuring the shift between two sample distributions

Question 41

what is the formula for the population shift index (PSI)

Answer 41

PSI=∑[(N_bin-B_bin )⋅ln⁡(N_bin/B_bin ) ]

Question 42

what values of PSI indicate: no significant shift, a minor shift, a significant shift

Answer 42

<0.1: no significant shift 0.1-0.25: minor shift >0.25: significant shift

Question 43

what is benchmarking

Answer 43

comparing a scorecard to an existing scorecard

Question 44

what is the KS statistic

Answer 44

the maximum difference between the CDFs of the distributions of defaults and non-defaults

Question 45

what is misclassification

Answer 45

the confusion matrix

Question 46

what is the ROC curve

Answer 46

the probability a randomly chosen non-default will be ranked righter than a randomly chosen default; plots the true positive rate against the false positive rate

Question 47

what is the formula of the accuracy ratio

Answer 47

AR=GINI/(Perfect GINI)

Question 48

what is the GINI index

Answer 48

the area between the Lorenz and random curve

Question 49

what is a lift chart

Answer 49

the cumulative % of defaults per decile divided by the total population % of defaults.

Question 50

what does weight of evidence do

Answer 50

transforms explanatory variables into a set of groups based on the similarity of the target variable distributions.

Question 51

what does WOE measure

Answer 51

how strong a group is at separating defaults from non-defaults

Question 52

what does a negative WOE signify?

Answer 52

more defaults than non-defaults

Question 53

what is the formula for WOE

Answer 53

WOE=ln⁡[((# non-defaults)/(total non-defaults))/((# defaults)/(total defaults))]

Question 54

what is a variable's information value

Answer 54

the predictive power of a single variable (its ability to separate defaults from non-defaults)

Question 55

what is the formula for information value

Answer 55

IV=∑[[(# non-defaults)/(total non-defaults)-(# defaults)/(total defaults)]⋅WOE_i

Question 56

what IV value ranges indicate: very weak weak moderate strong

Answer 56

<0.02: very weak 0.02-0.1: weak 0.1-0.3: moderate 0.3+: strong