Download the free Kindle app and start reading Kindle books instantly on your smartphone, tablet, or computer - no Kindle device required.
Read instantly on your browser with Kindle for Web.
Using your mobile phone camera - scan the code below and download the Kindle app.
Follow the author
Report an issue with this product or sellerThere is a newer edition of this item:
$49.99
This title will be released on October 9, 2026.
Customers who viewed this item also viewed
- Modern Time Series Forecasting with Python: Industry-ready machine learning and deep learning time series analysis with PyTorch and pandasPaperbackFREE Shipping by AmazonGet it as soon as Sunday, Oct 110% Claimed
From the Publisher
Key Features:
- Learn how to derive insights from time series and analyze a model’s performance
- Identify advantages and disadvantages of common time series models in machine learning
- Learn techniques such as autoencoders, InceptionTime, DeepAR, N-BEATS, Recurrent Neural Networks, ConvNets, and Informer
- Evaluate high-performance forecasting solutions
What are the key takeaways from this book?
This book teaches you how to analyze time series datasets with machine learning principles. An important takeaway is understanding how the ML landscape for time series has evolved. Readers will become aware of the tools for time series analysis. Each topic in the book provides a review of the latest research and an introduction to popular libraries with examples. Dedicated chapters focus on robust machine learning, deep learning, and reinforcement learning models for time series.
Major Topics:
- Probabilistic models for time series such as Facebook Prophet, Markov models, Fuzzy models, and counter-factual causal models such as Bayesian structural time series models as proposed by Google
- Multivariate forecasting and practical examples for multivariate multistep forecasts of energy demand with deep learning models
- Time series techniques such as bandit algorithms and Deep Q-Learning, and their application for a recommender system and trading algorithm
What trajectory does your book follow to help readers master time series with Python?
Opening a book and wondering "where do I begin?" can be overwhelming. This book starts by explaining concepts from the ground up. I’ve begun with a historical overview, a broad overview, and a basic introduction to Python for time series, which includes data loading and preprocessing.
My intention was to start with the basics and build your understanding in a step-by-step manner; the pace picks up gradually, with the complexity increasing with each chapter. Time series data manipulation, statistical methods, and time series analysis covered in earlier chapters are systematically connected to a repertoire of ML methods as the book takes you from loading time series datasets from various sources to understanding deep learning models.
Code samples help you to apply all methods to your own problems, and all notebooks used in this book come with links to Google Colab, enabling you to not just read about and learn the theory of new methods but also to experiment with them.
How does your book differ from other books on machine learning for time series?
In the last few years, a lot of progress has been made in machine learning for time series. Traditional methods such as ARIMA now face stiff competition from specialized methods for time series. While there are countless books on machine learning with Python and also a few on time series with Python, I haven’t seen any that include advancements in machine learning for time series within the last 15 years.
Furthermore, many books focus on traditional techniques, but not on recent machine learning approaches. Machine learning methods have won recent high-profile time series competitions such as M4 and M5, but these methods are not covered elsewhere. This book fills this gap.
Finally, while other books focus heavily on time series analysis, which is essential for some more traditional models, in this book, I present the best practices for machine learning workflows applied to time series and guide you through matching the right model to the right problem.
If you want to learn about time series and wish to transition from R to Python, you will find this book extremely useful as it includes several practical examples and applications.
Table of Contents:
- Preprocessing Time Series
- Forecasting with Moving Averages and Autoregressive Models
- Machine Learning Models for Time Series
- Online Learning for Time Series
- Probabilistic Models for Time Series
- Deep Learning for Time Series
- Reinforcement Learning for Time Series
- ...and more!
About the author
