Preface

Forecasting is the art and science of predicting the future, not with a magical crystal ball, but with proven statistical methods based on objective data. With more data and an improved understanding of forecasting methods, decisions become more influenced by forecasts.

We can forecast many things, from tomorrow’s weather to financial markets to election results. This book focuses on one specific use case for forecasting: forecasting demand for products or services.

There are many books and articles to learn from (see Chapter 22). This book is not an in-depth technical treatise. Instead, it aims to bring you up to speed with minimal pain, prepare you to learn more, or have intelligent conversations about forecasting with experts. Read the Preface to find out whether this book is for you.

This book is also available online at https://dfep.netlify.app/, where we will update it continually.

Happy forecasting!

Who this book is for

This book is a high-level introduction to demand forecasting. It will, by itself, not turn you into a forecaster. It gives you an overview of the most common forecasting methods and the mindset behind forecasting. This book is for you if you

  • Are a manager responsible for forecasters
  • Are an IT professional whose responsibilities include administering a forecasting system
  • Are an executive making decisions based on the forecasts generated by someone else in your work
  • Contemplate learning more about forecasting without delving deep into the details (yet)
  • Want to learn about forecasting and need a companion book to give you an overview, along with more technical in-depth materials

How to read this book

We divide this book into six parts:

  • The “Introduction” gives an overview, setting the scene and explaining the basic philosophy of this book: forecasting is indispensable for making decisions under uncertainty.
  • “Forecasting basics” discusses the essential toolbox of forecasters: knowing your time series, time series features, and time series decomposition into seasonal and trend components.
  • “Forecasting models” gives a very high-level introduction to standard models: simple methods, Exponential Smoothing, ARIMA, causal models, count data and intermittent demand, and Artificial Intelligence/Machine Learning.
  • “Forecasting quality” explains how to assess the quality of a forecast (which holds some pitfalls for the unwary) and what forecasting competitions are.
  • “Forecasting organization” discusses organizational and human aspects of forecasting: what to look for in a forecaster, how to deal with forecasters, how to build a forecasting team, and why forecasting fails.
  • Finally, “Learning more” points to various resources to learn more, from textbooks at various levels of technical detail, across organizations and events to websites.

While we did our best to provide good information and advice on forecasting, your forecasts are your own, so use them responsibly. If you create and act on a forecast, we disclaim any responsibility for any adverse outcomes. This book is a learning guide; it will not automatically make your forecasts as accurate as you would like them to be.

Acknowledgments

We want to thank Business Expert Press, publishers of our earlier book (Kolassa and Siemsen, 2016), for allowing us to reuse much of the material in this new book and, more generally, for being great publishers. Special thanks to Scott Isenberg for his guidance over the years.

In addition, we would like to thank CRC Press for allowing us to publish an online version of this book.

We thank Marjolein Poortvliet, who created a very inspirational title page for this book and endured our requests for changes and adaptations.

We are also grateful to the people who created and maintained the free tools we used in preparing this book, from the statistical programming environment R (R Core Team, 2022) and the packages we used (Arnold, 2021; Hyndman, 2020; Hyndman and Kourentzes, 2018; Hyndman et al., 2021; Schauberger and Walker, 2022; Slowikowski, 2023; Stoffer and Poison, 2023; Wickham et al., 2019; Xie, 2022; Zhu, 2021) to  (The LaTeX community, n.d.).

SK would like to express his gratitude to his long-suffering but patient family, Iris, Sophie and Philipp, who did not see much of him when this book was on the home stretch. He is also thankful to his co-authors for a long, sometimes grueling, but never boring journey together.

BRT would like to thank his partner, Maryam, for her love and continuous patience during the whole process of writing this book. He also expresses his gratitude to his co-authors for their valuable contributions and the opportunity to collaborate on this book.

ES deeply thanks his wife, Min, and children, Oskar and Thalia. Their love has always kept him going. ES also thanks his co-authors, his team, and Dean Samba for allowing him to work on this project despite managing a portfolio of graduate programs.

Finally, we are grateful to you, our reader, for investing your time and energy in reading our book. We hope you will learn a lot and even occasionally will be entertained!

References

Arnold, J. B. (2021). Ggthemes: Extra themes, scales and geoms for ’ggplot2’. R package version 4.2.4.
Hyndman, R. J. (2020). fpp2: Data for “Forecasting: Principles and Practice” (2nd Edition). R package version 2.4.
Hyndman, R. J., and Kourentzes, N. (2018). thief: Temporal HIErarchical Forecasting. R package version 0.3.
Hyndman, R. J., Lee, A., Wang, E., and Wickramasuriya, S. (2021). hts: Hierarchical and Grouped Time Series. R package version 6.0.2.
Kolassa, S., and Siemsen, E. (2016). Demand forecasting for managers. New York, NY: Business Expert Press.
R Core Team. (2022). R: A language and environment for statistical computing. Vienna, Austria: R Foundation for Statistical Computing.
Schauberger, P., and Walker, A. (2022). Openxlsx: Read, write and edit xlsx files. R package version 4.2.5.1.
Slowikowski, K. (2023). Ggrepel: Automatically position non-overlapping text labels with ’ggplot2’. R package version 0.9.3.
Stoffer, D., and Poison, N. (2023). astsa: Applied Statistical Time Series Analysis. R package version 2.0.
The LaTeX community. (n.d.). The LaTeX Project.
Wickham, H., Averick, M., Bryan, J., Chang, W., McGowan, L., François, R., … Yutani, H. (2019). Welcome to the tidyverse. Journal of Open Source Software, 4(43), 1686.
Xie, Y. (2022). bookdown: Authoring Books and Technical Documents with R Markdown. R package version 0.30.
Zhu, H. (2021). kableExtra: Construct complex table with ’kable’ and pipe syntax. R package version 1.3.4.