Professors Rob Hyndman and George Athanasopoulos together with their co-authors Azul Garza, Cristian Challu and Max Mergenthaler from Nixtla and Kin G Olivares from Amazon, are excited to announce the launch of Forecasting: Principles and Practice, the Pythonic Way, a long-awaited resource that brings the trusted content of https://otexts.com/fpp3/ in
Whether you’re teaching, learning, or applying forecasting in the real world, FPP for Python provides free, online access to:
- A practical, modern approach to time series forecasting.
- Fully reproducible Python code and examples
- Exercises and datasets ideal for teaching (instructor resources) or self-study
- Two new chapters covering recent advancements:
- Neural Networks: introduces deep learning methods for time series forecasting, including multilayer perceptrons, CNNs, and RNNs, as well as state-of-the-art models like NHITS. It provides insights into their training, and techniques for improving performance.
- Foundation Forecasting Models: explores cutting-edge approaches using large pre-trained transformers, discussing their theoretical basis, architectures, and applications
- The same accessible explanations trusted by 2,100+ instructors worldwide
With over 10,000 citations, 50M+ pageviews, and translations in multiple languages (Chinese, Korean, Japanese, Greek, Italian, Portuguese, Russian), Forecasting: Principles and Practice has become a global standard – and it’s available to the Python community.
Explore now: https://otexts.com/fpppy