Are you tired of sifting through multiple resources to learn about time series forecasting? Look no further! My book, Mastering Modern Time Series Forecasting, has been holding the #1 spot on Leanpub in Machine Learning, Forecasting, and Time Series categories for weeks now.
Trusted by readers in over 100 countries, this book covers everything you need to know about time series forecasting, from classical methods like ARIMA, SARIMA, and Prophet to modern ML/DL models like LightGBM, N-BEATS, TFT, and Transformers. The book takes a Python-first approach, using popular libraries like scikit-learn, PyTorch, statsmodels, and Darts, to ensure that the concepts are production-ready and scalable.
What sets this book apart is its practical focus on real-world case studies, handling messy data, feature engineering, and robust evaluation. It also delves into explainability and uncertainty, including SHAP values, conformal prediction, backtesting, and model confidence bands.
The best part? This is a living book with free lifetime updates, so you’ll get access to new chapters and updates as they’re added.
I wrote this book because I couldn’t find a single resource that balanced theory, practice, and production concerns. I hope it saves you months of trial-and-error and becomes your go-to resource for time series forecasting.
Do you have any questions or feedback about the book? I’d be happy to discuss any chapter or topic in more depth.