## Credit Risk Modeling for Basel/IFRS 9 using R/Python/SAS

##### 📅 October 27-28

🌍 English

## About This Course

In this course, students learn how to develop credit risk models in the context of the Basel and IFRS 9 guidelines. The course extensively reviews the 3 key credit risk parameters: Probability of Default (PD), Loss Given Default (LGD), and Exposure at Default (EAD). Modeling methods, performance measurement and benchmarks are discussed into great detail. The course provides a sound mix of both theoretical and technical insights, as well as practical implementation details. These are illustrated by several real-life case studies and examples. The course also features code examples in R, Python and SAS. Throughout the course, the instructors also extenisvely report upon their research and industry experience. A certificate signed by the instructors is provided upon successful completion.

## Requirements

Before subscribing to this course, you should have business expertise in credit risk and a basic understanding of descriptive statistics (e.g., mean, median, standard deviation, etc.) and inference (e.g., confidence intervals, hypothesis testing). Previous R, Python and SAS experience is helpful but not necessary.

## Course Outline

- Introduction
- Instructor team
- Our Credit Risk Publications
- Software
- R/Python tutorials
- Data sets
- Disclaimer
- Credit scoring
- Introduction
- Retail credit scoring
- Application scoring
- Behavioral scoring
- Corporate credit scoring
- Prediction approach
- Expert-based approach
- Agency Rating approach
- Shadow Ratings approach
- Quiz
- Discussion
- Basel Accords/IFRS 9/CECL
- Regulatory versus Economic Capital
- Basel I and II Capital Accords
- Basel III
- Basel IV
- Basel approaches to model credit risk
- Standardized approach
- IRB approach
- Default rating specifics
- Basel IRB model architecture
- Risk weight functions for retail
- Merton model
- Risk weight functions
- IFRS 9
- Credit risk model lifecycle
- Quiz
- Discussion
- Data Preprocessing
- Motivation
- Types of data
- Types of variables
- Denormalizing data
- Sampling
- Visual data exploration
- Descriptive statistics
- Missing values
- Outliers
- Categorization
- WOE and IV
- Variable Transformation
- Quiz
- Classification Techniques
- Linear Regression
- Logistic Regression
- Nomograms
- Decision trees
- Cumulative logistic regression
- Multiclass decision trees
- Quiz
- Measuring the performance of credit scoring classification models
- Split sample method
- Cross-validation
- Single sample method
- Confusion matrix (classification accuracy, classification error, sensitivity, specificity)
- ROC curve and area under ROC curve
- CAP curve and Accuracy Ratio
- Lift curve
- Kolmogorov-Smirnov distance
- Mahalanobis distance
- Performance benchmarks
- Multiclass confusion matrix
- Notch difference graph
- Multiclass AUC
- Quiz
- Survival analysis
- Censoring
- Time varying covariates
- Survival distributions
- Kaplan Meier analysis
- Accelerated Failure Time models
- Proportional hazards model
- Discrete survival analysis
- Competing risks
- Mixture cure modeling
- Survival analysis in credit risk modeling
- Evaluating survival analysis
- Quiz
- Defining default ratings and calibrating PD
- Defining default ratings
- Rating migration analysis
- Rating philosophy
- Rating mobility
- PD calibration
- Quiz
- Modeling Loss Given Default (LGD)
- Definition of default
- Definition of LGD
- Ways of measuring LGD
- LGD according to Basel
- Constructing an LGD data set
- Complete business Cycle
- Default definition and cures
- LGD measurement
- Length of workout period
- Incomplete workouts
- Discount rate
- LGD < 0 or LGD>100%
- Indirect costs
- LGD drivers
- Data preprocessing
- Challenges in LGD modeling
- LGD modeling approaches
- Segmentation
- Regression trees
- Linear regression LGD modeling
- Beta regression LGD modeling
- Logistic regression LGD modeling
- Cumulative logistic regression LGD modeling
- Two stage LGD models
- Advanced LGD models
- Performance measures
- LGD ratings
- LGD calibration
- Exposures in default
- Modeling Exposure At Default (EAD)
- Defining Exposure At Default (EAD)
- EAD according to Basel
- Drawings post default
- EAD Development Sample
- CCF < 0 or CCF>100%
- EAD modeling
- EAD Case Study
- Correlation between PD/LGD/EAD

##### 👩🏫 Lecturers

Prof. dr. Wouter Verbeke

Professor at KU Leuven

Prof. dr. Bart Baesens

Professor at KU Leuven

##### 🏢 Location

Grote Markt 5, 3000 Leuven, Belgium

##### 🏫 Organizer

##### 💼 Register

This course is in the past, registration is no longer possible.

### Price and Registration

This course is in the past, registration is no longer possible.