English | MP4 | AVC 1280×720 | AAC 44KHz 2ch | 75 lectures (6h 51m) | 2.19 GB
A complete data science case study: preprocessing, modeling, model validation and maintenance in Python
Hi! Welcome to Credit Risk Modeling in Python. The only online course that teaches you how banks use data science modeling in Python to improve their performance and comply with regulatory requirements. This is the perfect course for you, if you are interested in a data science career. Here’s why:
The instructor is a proven expert (PhD from the Norwegian Business school, who has taught in world renowned universities such as HEC, the University of Texas, and the Norwegian Business school).
The course is suitable for beginners. We start with theory and initial data pre-processing and gradually solve a complete exercise in front of you
Everything we cover is up-to-date and relevant in today’s development of Python models for the banking industry
This is the only online course that shows the complete picture in credit risk in Python (using state of the art techniques to model all three aspects of the expected loss equation – PD, LGD, and EAD) including creating a scorecard from scratch
Here we show you how to create models that are compliant with Basel II and Basel III regulations that other courses rarely touch upon
We are not going to work with fake data. The dataset used in this course is an actual real-world example
You get to differentiate your data science portfolio by showing skills that are highly demanded in the job marketplace
What is most important – you get to see first-hand how a data science task is solved in the real-world
Most data science courses cover several frameworks, but skip the pre-processing and theoretical part. This is like learning how to taste wine before being able to open a bottle of wine.
We don’t do that. Our goal is to help you build a solid foundation. We want you to study the theory, learn how to pre-process data that does not necessarily come in the ‘’friendliest’’ format, and of course, only then we will show you how to build a state of the art model and how to evaluate its effectiveness.
Throughout the course, we will cover several important data science techniques.
- Weight of evidence
- Information value
- Fine classing
- Coarse classing
- Linear regression
- Logistic regression
- Area Under the Curve
- Receiver Operating Characteristic Curve
- Gini Coefficient
- Kolmogorov-Smirnov
- Assessing Population Stability
- Maintaining a model
What you’ll learn
- Improve your Python modeling skills
- Differentiate your data science portfolio with a hot topic
- Fill up your resume with in demand data science skills
- Build a complete credit risk model in Python
- Impress interviewers by showing practical knowledge
- How to preprocess real data in Python
- Learn credit risk modeling theory
- Apply state of the art data science techniques
- Solve a real-life data science task
- Be able to evaluate the effectiveness of your model
- Perform linear and logistic regressions in Python
Table of Contents
Introduction
What does the course cover
What is credit risk and why is it important
Expected loss EL and its components PD LGD and EAD
What is credit risk and why is it important
Capital adequacy regulations and the Basel II accord
Expected loss EL and its components PD LGD and EAD
Basel II approaches SA FIRB and AIRB
Capital adequacy regulations and the Basel II accord
Basel II approaches SA F
Different facility types asset classes and credit risk modeling approaches
Different facility types asset classes and credit risk modeling approaches
Setting up the working environment
Setting up the environment
Why Python and why Jupyter
Installing Anaconda
Jupyter Dashboard
Jupyter Dashboard
Installing the sklearn package
Dataset description
Our example consumer loans A first look at the dataset
Dependent variables and independent variables
Our example consumer loans A first look at the dataset
Dependent variables and independent variables
General preprocessing
Importing the data into Python
Preprocessing few continuous variables
Preprocessing few discrete variables
Check for missing values and clean
Importing the data into Python
Preprocessing few continuous variables
Preprocessing few continuous variables Homework
Preprocessing few discrete variables
Check for missing values and clean
Check for missing values and clean Homework
PD Model Data Preparation
How is the PD model going to look like
Dependent variable Good Bad default definition
Fine classing weight of evidence and coarse classing
Information value
Data preparation Splitting data
Data preparation An example
Data preparation Preprocessing discrete variables automating calculations
Data preparation Preprocessing discrete variables creating dummies Part 1
Data preparation Preprocessing discrete variables creating dummies Part 2
Data preparation Preprocessing continuous variables Automating calculations
How is the PD model going to look like
Data preparation Preprocessing continuous variables creating dummies Part 1
Dependent variable Good Bad default definition
Data preparation Preprocessing continuous variables creating dummies Part 2
Fine classing weight of evidence and coarse classing
Data preparation Preprocessing continuous variables creating dummies Part 3
Information value
Data preparation Splitting data
Data preparation An example
Data preparation Preprocessing discrete variables automating calculations
Data preparation Preprocessing discrete variables visualizing results
Data preparation Preprocessing discrete variables creating dummies Part 1
Data preparation Preprocessing discrete variables creating dummies Part 2
Data preparation Preprocessing discrete variables Homework
Data preparation Preprocessing continuous variables Automating calculations
Data preparation Preprocessing continuous variables creating dummies Part 1
Data preparation Preprocessing continuous variables creating dummies Part 2
Data preparation Preprocessing continuous variables creating dummies Homework
Data preparation Preprocessing continuous variables creating dummies Part 3
Data preparation Preprocessing continuous variables creating dummies Homework
Data preparation Preprocessing the test dataset
PD model data preparation notebooks
PD model estimation
The PD model Logistic regression with dummy variables
Build a logistic regression model with pvalues
Interpreting the coefficients in the PD model
The PD model Logistic regression with dummy variables
Loading the data and selecting the features
PD model estimation
Build a logistic regression model with p
Interpreting the coefficients in the PD model
PD model validation
Outofsample validation test
Evaluation of model performance accuracy and area under the curve AUC
Evaluation of model performance Gini and KolmogorovSmirnov
Out
Evaluation of model performance accuracy and area under the curve AUC
Evaluation of model performance Gini and Kolmogorov
Applying the PD Model for decision making
Creating a scorecard
Calculating credit score
From credit score to PD
Setting cutoffs
Calculating probability of default for a single customer
Creating a scorecard
Calculating credit score
From credit score to PD
Setting cut
Setting cutoffs Homework
PD model logistic regression notebooks
PD model monitoring
PD model monitoring via assessing population stability
Population stability index calculation and interpretation
PD model monitoring via assessing population stability
Population stability index preprocessing
Population stability index calculation and interpretation
Homework building an updated PD model
LGD and EAD Models Preparing the data
LGD and EAD models independent variables
LGD and EAD models dependent variables
LGD and EAD models distribution of recovery rates and credit conversion factors
LGD and EAD models independent variables
LGD and EAD models dependent variables
LGD and EAD models distribution of recovery rates and credit conversion factors
LGD model
LGD model testing the model
LGD model stage 2 linear regression with comments
LGD model stage 2 linear regression evaluation
LGD model combining stage 1 and stage 2
LGD model preparing the inputs
LGD model testing the model
LGD model estimating the accuracy of the model
LGD model saving the model
LGD model stage 2 linear regression
LGD model stage 2 linear regression evaluation
LGD model combining stage 1 and stage 2
Homework building an updated LGD model
EAD model
EAD model estimation and interpretation
EAD model validation
EAD model estimation and interpretation
EAD model validation
Homework building an updated EAD model
Calculating expected loss
Calculating expected loss
Calculating expected loss
Homework calculate expected loss on more recent data
Completing 100
Resolve the captcha to access the links!