This book is still in progress and the code might change before the full release in Spring 2022

If you do not have the book yet, make sure to grab a copy here

In this book, you learn how to build predictive models for time series. Both the statistical and deep learnings techniques are covered, and the book is 100% in Python!

Specifically, you will learn how to:

• Recognize a time series forecasting problem and build a performant predictive model
• Create univariate forecasting models that accound for seasonality and external variables
• Build multivariate forecasting models to predict many time series at once
• Leverage large datasets using deep learning models for forecasting (implementation in TensorFlow/Keras)
• Automate the forecasting process

Plus, the book comes with a ton of hands-on projects with real-life data, such as the earnings per share of Johnson & Johnson, the daily stock price of Google, the US macroeconomic data, the volume of antidiabetic drug prescription in Australia, and much more.

Get your copy now!

I highly recommend that you read the book and code along. This is the best way to take the most out of the book.

Each folder corresponds to a chapter. They each contain the notebook with all the code presented in that chapter. The code is in order of appearance in the book. When appropriate, there is also a data folder cointaining the CSV file used in that chapter.

Keep in mind that changes might be done anytime before the final release.

The following chapters are accessible but might still be modified before the final release. That's why your feedback is important, so we can improve the book together.

• Ch 1: Understanding time series forecasting
• Ch 2: A naïve prediction of the future
• Ch 3: Going on a random walk
• Ch 4: Modeling a moving average process
• Ch 5: Modeling an autoregressive process

• Ch 6: Modeling complex time series
• Ch 7: Forecasting non-stationary time series
• Ch 8: Accounting for seasonality
• Ch 9: Adding external variables to our model
• Ch 10: Forecasting multiple time series
• Ch 11: Captonse project - Forecasting the number of antidiabetic drug prescriptions in Australia

• Ch 12: Introducing deep learning for time series forecasting
• Ch 13: Data windowing and creating baselines for deep learning
• Ch 14: Baby steps with deep learning
• Ch 15: Remembering the past with LSTMs
• Ch 16: Filtering our time series with CNN
• Ch 17: Predicting the difference with residual networks
• Ch 18: Using our predictions to make more predictions (autoregressive LSTM)
• Ch 19: Capstone project - Predicting the electricity consumptions of households
• Ch 20: Working with Prophet for automated time series forecasting
• Ch 21: Capstone project - Forecasting beer production in Australia