
Develop a behavioral scorecard using past payment behavior, demographic information, and credit bureau status to predict one-year default risk with logistic regression for existing personal loan customers.
Explore sample data fields used to build a retail portfolio scorecard, including demographics, financials, collateral, credit bureau status, and payment behavior to predict loan default.
Develop WOE and IV values in SAS using a bivariate macro to compute default totals and rates, then interpret and screen variables for credit risk modeling.
Learn how to validate a logistic regression model through rank ordering across deciles, and interpret KS statistics, the Gini coefficient, and the Lorenz curve to assess discriminatory power.
Compute the Brier score to assess calibration by averaging the squared differences between predicted probabilities and actual outcomes across bins. A lower score signals a more accurate credit risk model.
Credit Risk Modeling is a technique used by lenders to determine the level of credit risk associated with extending credit to a borrower. In other words, it’s a tool to understand the credit risk of a borrower. This is especially important because this credit risk profile keeps changing with time and circumstances. Credit risk modeling is the process of using statistical techniques and machine learning to assess this risk. The models use past data and various other factors to predict the probability of default and inform credit decisions.
This course teaches you how banks use statistical modeling in SAS to prepare credit risk scorecard which will assist them to predict the likelihood of default of a customer. We will deep dive into the entire model building process which includes data preparation, scorecard development and checking for a robust model, model validation and checking for the accuracy of the model step by step from scratch. This course covers the following in detail with output interpretation, best practices and SAS Codes explanations :
1) Understanding the dataset and the key variables used for scorecard building
2) Development sample exclusions
3) Observation and Performance window
4) Model Design Parameters
5) Vintage and Roll Rate Analysis
6) Data Preparation which includes missing values and outlier identification and treatment
7) Bifurcating Training and Test datasets
8) Understanding the dataset in terms of key variables and data structure
9) Fine and Coarse classing
10) Information value and WOE
11) Multicollinearity
12) Logistic Regression Model development with statistical interpretation
13) Concordance, Discordance, Somer's D and C Statistics
14) Rank Ordering, KS Statistics and Gini Coefficient
15) Checking for Clustering
16) Goodness of fit test
17) Model Validation and
18) Brier Score for model accuracy