Skip to main content

Course

GARCH Models in Python

IntermediateSkill Level

4.8+

Updated 06/2022

Learn about GARCH Models, how to implement them and calibrate them on financial data from stocks to foreign exchange.

Start Course for Free

PythonApplied Finance

4 hr

15 videos

54 Exercises

3,950 XP

10,602

Statement of Accomplishment

Loved by learners at thousands of companies

Training a Team?

Try for Business

Course Description

Volatility is an essential concept in finance, which is why GARCH models in Python are a popular choice for forecasting changes in variance, specifically when working with time-series data that are time-dependant. This course will show you how and when to implement GARCH models, how to specify model assumptions, and how to make volatility forecasts and evaluate model performance. Using real-world data, including historical Tesla stock prices, you’ll gain hands-on experience of how to better quantify portfolio risks, through calculations of Value-at-Risk, covariance, and stock Beta. You’ll also apply what you’ve learned to a wide range of assets, including stocks, indices, cryptocurrencies, and foreign exchange, preparing you to go forth and use GARCH models.

Prerequisites

Time Series Analysis in Python

1

GARCH Model Fundamentals

What are GARCH models, what are they used for, and how can you implement them in Python? After completing this first chapter you’ll be able to confidently answer all these questions.

Why do we need GARCH models

Understand volatility

Observe volatility clustering

Calculate volatility

What are ARCH and GARCH

Review GARCH model basics

Simulate ARCH and GARCH series

Observe the impact of model parameters

How to implement GARCH models in Python

Review "arch" documentation

Implement a basic GARCH model

Make forecast with GARCH models

2

GARCH Model Configuration

A normal GARCH model is not representative of the real financial data, whose distributions frequently exhibit fat tails, skewness, and asymmetric shocks. In this chapter, you’ll learn how to define better GARCH models with more realistic assumptions. You’ll also learn how to make more sophisticated volatility forecasts with rolling window approaches.

Distribution assumptions

Fat tails and skewness

Plot distribution of standardized residuals

Fit a GARCH with skewed t-distribution

Mean model specifications

Check mean model assumptions

Effect of mean model on volatility predictions

Volatility models for asymmetric shocks

Modeling of asymmetric responses of volatility

Fit GARCH models to cryptocurrency

Compare GJR-GARCH with EGARCH

GARCH rolling window forecast

Why use rolling window forecast

Fixed rolling window forecast

Compare forecast results

3

Model Performance Evaluation

This chapter introduces you to the KISS principle of data science modeling. You’ll learn how to use p-values and t-statistics to simplify model configuration, use ACF plot, Ljung-Box test to verify model assumptions and use likelihood and information criteria for model selection.

Significance testing of model parameters

Keep it simple stupid

Simplify the model with p-values

Simplify the model with t-statistics

Validation of GARCH model assumptions

Detect autocorrelations

Ljung-Box test

Goodness of fit measures

Goodness of fit basics

Pick a winner based on log-likelihood

Pick a winner based on AIC/BIC

GARCH model backtesting

Backtesting basics

Backtesting with MAE, MSE

4

GARCH in Action

In this final chapter, you’ll learn how to apply the GARCH models you’ve previously learned to practical financial world scenarios. You’ll develop your skills as you become more familiar with VaR in risk management, dynamic covariance in asset allocation, and dynamic Beta in portfolio management.

VaR in financial risk management

VaR concept

Compute parametric VaR

Compute empirical VaR

Dynamic covariance in portfolio optimization

Covariance concept

Compute GARCH covariance

Compute dynamic portfolio variance

Dynamic Beta in portfolio management

Beta concept

Compute dynamic stock Beta

Congratulations!

GARCH Models in Python

Course
Complete

Earn Statement of Accomplishment

Add this credential to your LinkedIn profile, resume, or CV
Share it on social media and in your performance reviewEnroll Now

Don’t just take our word for it

*4.8

from 185 reviews

86%

14%

1%

0%

0%

Sort by

Christian

last week

Rafael de Jesús

2 weeks ago

hubert

2 weeks ago

Barbara

2 weeks ago

Alessandro

2 weeks ago

Himan

2 weeks ago

Christian

Rafael de Jesús

hubert

FAQs

What financial datasets are used in this course?

You work with real-world data including historical Tesla stock prices, stock indices, cryptocurrencies, and foreign exchange rates to practice building and evaluating GARCH models.

What can I do with GARCH models after completing this course?

You will be able to forecast volatility, calculate Value-at-Risk for risk management, estimate dynamic covariance for asset allocation, and compute dynamic Beta for portfolio management.

Do I need time series analysis experience before enrolling?

Yes. The prerequisites include Time Series Analysis in Python and Manipulating Time Series Data in Python, along with pandas and Intermediate Python.

Does the course cover model selection and evaluation?

Yes. Chapter 3 teaches you to use p-values, t-statistics, ACF plots, the Ljung-Box test, and information criteria like AIC and BIC to select and validate your models.

How does this course handle real-world features like fat tails and asymmetric shocks?

Chapter 2 covers configuring GARCH models with more realistic distribution assumptions, including fat-tailed and skewed distributions, plus rolling window forecasting approaches.

Join over 19 million learners and start GARCH Models in Python today!

Grow your data skills with DataCamp for Mobile

Make progress on the go with our mobile courses and daily 5-minute coding challenges.