front matter

 

preface

Working at a bank, I quickly realized how time is an important factor. Interest rates vary over time, people’s spending varies over time, asset prices vary over time. Yet I found most people, including me, were uncomfortable with time series. So I decided to learn time series forecasting.

It turned out to be harder than expected because every resource I found was in R. I am comfortable with Python, and Python is undoubtedly the most popular language for data science in the industry. While R constrains you to statistical computing, Python allows you to code websites, perform machine learning, deploy models, build servers, and more. Therefore, I had to translate a lot of R code into Python to learn time series forecasting. That’s when I recognized the gap, and I was lucky enough to be given the opportunity to write a book about it.

With this book, I hope to create a one-stop reference for time series forecasting with Python. It covers both statistical and machine learning models, and it also discusses automated forecasting libraries, as they are widely used in the industry and often act as baseline models. This book greatly emphasizes a hands-on, practical approach, with various real-life scenarios. In real life, data is messy, dirty, and sometimes missing, and I wanted to give readers a safe space to experiment with those difficulties, learn from them, and easily transpose those skills into their own projects.

acknowledgments

about this book

Who should read this book?

How this book is organized: A roadmap

About the code

liveBook discussion forum

Author online

about the author

about the cover illustration