Online Course top-rated
Credit Risk Modeling in Python

Name: Credit Risk Modeling in Python Course
Price: 36 USD

Blend credit risk modeling skills with Python programming: Learn how to estimate a bank’s loan portfolio's expected loss

4.8

862 reviews on

Start for free

10,507 students already enrolled

Skill level:

Advanced

Duration:

8 hours

Lessons (7 hours)

Practice exams (1.75 hours)

CPE credits:

12

CPE stands for Continuing Professional Education and represents the mandatory credits a wide range of professionals must earn to maintain their licenses and stay current with regulations and best practices. One CPE credit typically equals 50 minutes of learning. For more details, visit NASBA's official website: www.nasbaregistry.org

Accredited

certificate

What you learn

Master retail banking value drivers and credit risk modeling fundamentals.
Understand credit risk modeling concepts like PD, LGD, EAD, and Basel II.
Apply logistic regression in Python for accurate credit risk prediction.
Boost your data cleaning and processing skills with real-world loan data.
Acquire specialized credit risk modeling skills to enhance your resume.
Become invaluable for data scientist roles in the retail banking sector.

Topics & tools

TheoryPythonData AnalysisProgrammingCredit RiskLogistic RegressionData PreprocessingFinance SkillsIndustry Specialization

Your instructor

Nikolay Georgiev

Data Science Leader | Business Impact with AI & Behavioral Insights

Track record

Bringing real-world expertise from leading global companies

Academic background

Doctorate (PhD), Economics & Business Administration

Media and recognition

Author of 11 scientific publications
Presenter at international conferences and colloquiums
Silver medalist in multiple academic competitions (Mathematics & Econometrics)

Ask Nikolay a question

Course overview

Accredited certificates

Course OVERVIEW

Description

CPE Credits: 12 Field of Study: Information Technology
Delivery Method: QAS Self Study

Credit risk modeling is the place where data science and fintech meet. It is one of the most important activities conducted in a bank, with the most attention since the recession. At present, it is the only comprehensive credit risk modeling course in Python available online – taking you from preprocessing, through probability of default (PD), loss given default (LGD) and exposure at default (EAD) modeling, all the way to calculating expected loss (EL).

Prerequisites

Python (version 3.8 or later), pandas library, and a code editor or IDE (e.g., Jupyter Notebook, Spyder, or VS Code)
Basic familiarity with Python programming is required.
Familiarity with NumPy is helpful but not mandatory.

Advanced preparation

Introduction to Python

Curriculum

68 lessons 80 exercises 5 exams

1. Introduction

38 min

We start by explaining why credit risk is important for financial institutions. We also define ground 0 terms, such as expected loss, probability of default, loss given default and exposure at default.

38 min

We start by explaining why credit risk is important for financial institutions. We also define ground 0 terms, such as expected loss, probability of default, loss given default and exposure at default.

What does the course cover Free

What is credit risk and why is it important? Free

Exercise Free

Expected loss (EL) and its components: PD, LGD and EAD Free

Exercise Free

Capital adequacy, regulations, and the Basel II accord Free

Exercise Free

Basel II approaches: SA, F-IRB, and A-IRB Free

Exercise Free

Different facility types (asset classes) and credit risk modeling approaches Free

Exercise Free
2. Setting up the environment

2 min

Here you will learn how to set up Python 3 and load up Jupyter. We’ll also show you what the Anaconda Prompt is and how you can use it to download and import new modules.

2 min

Here you will learn how to set up Python 3 and load up Jupyter. We’ll also show you what the Anaconda Prompt is and how you can use it to download and import new modules.

Setting up the environment Free

Installing the relevant packages Free
3. Dataset description

9 min

Our example focuses on consumer loans. Since there are more than 100 potential features, we've devoted a complete section to explain why some features are chosen over others.

9 min

Our example focuses on consumer loans. Since there are more than 100 potential features, we've devoted a complete section to explain why some features are chosen over others.

Our example: consumer loans. A first look at the dataset Free

Exercise Free

Dependent variables and independent variables Free

Exercise Free
4. General preprocessing

29 min

Each raw datasets has its drawbacks. While most preprocessing is model specific, in some cases (like missing values imputation), we could generalize the data preparation.

29 min

Each raw datasets has its drawbacks. While most preprocessing is model specific, in some cases (like missing values imputation), we could generalize the data preparation.

Importing the data into Python Free

Exercise Free

Preprocessing few continuous variables Free

Preprocessing few continuous variables Homework Free

Exercise Free

Preprocessing few discrete variables Free

Exercise Free

Check for missing values and clean Free

Check for missing values and clean Homework Free

Exercise Free
5. PD model: data preparation

117 min

Once we have completed all general preprocessing, we dive into model-specific preprocessing. We employ fine classing, coarse classing, weight of evidence and information value criterion to achieve the probability of default preprocessing. Conventionally, we should turn all variables into dummy indicators prior to modeling.

117 min

Once we have completed all general preprocessing, we dive into model-specific preprocessing. We employ fine classing, coarse classing, weight of evidence and information value criterion to achieve the probability of default preprocessing. Conventionally, we should turn all variables into dummy indicators prior to modeling.

How is the PD model going to look like?

Exercise

Dependent variable: Good/ Bad (default) definition

Exercise

Constructing independent variables

Exercise

Information value

Exercise

Data preparation. Splitting data

Exercise

Data preparation. Preprocessing one discrete variable

Exercise

Data preparation. Preprocessing discrete variables: automating calculations

Exercise

Data preparation. Preprocessing discrete variables: visualizing results

Data Preparation. Preprocessing Discrete Variables: Creating Dummies (part 1)

Exercise

Data Preparation. Preprocessing Discrete Variables: Creating Dummies (part 2)

Data Preparation. Preprocessing Discrete Variables: Creating Dummies (part 2)

Exercise

Data preparation. Preprocessing continuous variables: automating calculations

Exercise

Data preparation. Preprocessing continuous variables: creating dummies (part 1)

Exercise

Data preparation. Preprocessing continuous variables: creating dummies (part 2)

Data preparation. Preprocessing continuous variables: creating dummies (part 2)

Exercise

Data preparation. Preprocessing continuous variables: creating dummies (part 3)

Creating dummies Homework

Exercise

Data preparation. Preprocessing the test dataset

Practice exam
6. PD model estimation

35 min

Having set up all variables to be dummies, we estimate the probability of default. The most intuitive and widely accepted approach is to employ a logistic regression.

35 min

Having set up all variables to be dummies, we estimate the probability of default. The most intuitive and widely accepted approach is to employ a logistic regression.

The PD model. Logistic regression with dummy variables

Exercise

Loading the data and selecting the features

PD model estimation

Build a logistic regression model with p-values.

Exercise

Interpreting the coefficients in the PD model

Exercise
7. PD model validation (test)

28 min

Since each model overfits the training data, it is crucial to test the results on out-of-sample observations. Consequently, we find its accuracy, its area under the curve (AUC), the Gini coefficient and the Kolmogorov-Smirnov test.

28 min

Since each model overfits the training data, it is crucial to test the results on out-of-sample observations. Consequently, we find its accuracy, its area under the curve (AUC), the Gini coefficient and the Kolmogorov-Smirnov test.

Out-of-sample validation (test).

Exercise

Evaluation of model performance: accuracy and area under the curve (AUC)

Exercise

Evaluation of model performance: Gini and Kolmogorov-Smirnov.

Exercise
8. Applying the PD model for decision making

37 min

In practice, banks don't really want a complicated Python-implemented model. Instead, they prefer a simple score-card which contains only yes and no questions that could be employed by any bank employee. In this section, we learn how to create one.

37 min

In practice, banks don't really want a complicated Python-implemented model. Instead, they prefer a simple score-card which contains only yes and no questions that could be employed by any bank employee. In this section, we learn how to create one.

Calculating probability of default for a single customer

Creating a scorecard

Exercise

Calculating credit score

Exercise

From credit score to PD

Exercise

Setting cut-offs

Setting cut-offs Homework

Practice exam

Exercise
9. PD model monitoring

29 min

Model estimation is extremely important, but an often-neglected step is model maintenance. A common approach is to monitor the population stability over time using the population stability index (PSI) and revisit our model if needed.

29 min

Model estimation is extremely important, but an often-neglected step is model maintenance. A common approach is to monitor the population stability over time using the population stability index (PSI) and revisit our model if needed.

PD model monitoring via assessing population stability

Exercise

Population stability index: preprocessing

Population stability index: calculation and interpretation

Population stability index: calculation and interpretation Homework

Practice exam

Exercise
10. LGD and EAD models

17 min

To calculate the final expected loss, we need three ingredients: probability of default (PD), loss given default (LGD) and exposure at default (EAD). In this section, we preprocess our data to be able to estimate the LGD and EAD models.

17 min

To calculate the final expected loss, we need three ingredients: probability of default (PD), loss given default (LGD) and exposure at default (EAD). In this section, we preprocess our data to be able to estimate the LGD and EAD models.

LGD and EAD models: independent variables

Exercise

LGD and EAD models: dependent variables

Exercise

LGD and EAD models: distribution of recovery rates and credit conversion factors

Practice exam

Exercise
11. LGD model

29 min

LGD models are often estimated using a beta regression. To keep the modeling part simpler, we employ a two-step regression model, which aims to simulate a beta regression. We combine the predictions from a logistic regression with those from a linear regression to estimate the loss given default.

29 min

LGD models are often estimated using a beta regression. To keep the modeling part simpler, we employ a two-step regression model, which aims to simulate a beta regression. We combine the predictions from a logistic regression with those from a linear regression to estimate the loss given default.

LGD model: preparing the inputs

LGD model: testing the model

Exercise

LGD model: estimating the accuracy of the model

LGD model: saving the model

LGD model: stage 2 – linear regression

Exercise

LGD model: stage 2 – linear regression evaluation

Exercise

LGD model: combining stage 1 and stage 2

LGD model: combining stage 1 and stage 2 Homework

Exercise
12. EAD model

11 min

The exposure at default (EAD) modeling is very similar to the LGD one. In this section, we take advantage of a linear regression to calculate EAD.

11 min

The exposure at default (EAD) modeling is very similar to the LGD one. In this section, we take advantage of a linear regression to calculate EAD.

EAD model estimation and interpretation

Exercise

EAD model validation

EAD model validation

Exercise
13. Calculating expected loss

17 min

After having calculated PD, LGD, and EAD, we reach the final step: computing expected loss (EL). This is also the number which is most interesting to C-level executives and is the finale of the credit risk modeling process.

17 min

After having calculated PD, LGD, and EAD, we reach the final step: computing expected loss (EL). This is also the number which is most interesting to C-level executives and is the finale of the credit risk modeling process.

Calculating expected loss

Calculating expected loss Homework

Exercise
14. Course exam

75 min

75 min

Course exam

Free lessons

1.1 What does the course cover

5 min

1.2 What is credit risk and why is it important?

5 min

1.3 Expected loss (EL) and its components: PD, LGD and EAD

4 min

1.4 Capital adequacy, regulations, and the Basel II accord

5 min

1.5 Basel II approaches: SA, F-IRB, and A-IRB

10 min

1.6 Different facility types (asset classes) and credit risk modeling approaches

9 min

Start for free

My background is industrial engineering. And I started my professional journey in sales. But I wanted a career that aligned with what I loved—technology and programming. 365 gave me everything: the structure, the mentorship, and the practical tools to switch. I started with math and stats, then moved on to SQL, Python, and building dashboards and apps at work. Today, I create tools that improve decision-making and manage data projects full-time—and I couldn’t be happier.

Pedro A.

Before 365: Industrial Engineer | NIBOL

After 365: Data Manager | NIBOL

See all reviews

94%

of AI and data science graduates

successfully change

or advance their careers.

I'm happy to say after about 6 months, I was getting job interviews as either a business analyst or data analyst, and now I'm working downtown at a tech company as a data analyst. 365 helped me level up my Excel, SQL, and data visualization skills—and start building real projects for my employer. After years in the lab, 365 Data Science finally found the right way to help me change to AI and data science, and I'm excited for what’s next.

Lee McFarling

Before 365: Lab technician | CDPHE

After 365: Data Analyst | Astreya

See all reviews

96%

of our students recommend

365 Data Science.

One of the biggest takeaways from the bootcamp was the ability to tackle real-world data problems. During my job interview, I was given a practical task by the hiring managers, and thanks to the skills I gained from 365 Data Science, I was the only candidate who solved it! After this, I was hired as a BI developer and have now moved up to a data engineer.