Product guide

Modern Time Series Forecasting with Python, 2nd Edition

Modern Time Series Forecasting with Python is a second-edition guide to practical forecasting, spanning foundational methods through machine learning and deep learning. Its scope includes feature engineering, evaluation, global models, ensembles, transformers and probabilistic forecasting. The listing recommends familiarity with Python and basic machine learning.

At a glance

Covers introductory time-series concepts, ARIMA and exponential smoothing before progressing to machine-learning and deep-learning methods.
Topics include preprocessing, feature engineering, model evaluation, global forecasting, ensembling and stacking.
Advanced areas include RNNs, transformers, N-BEATS, conformal prediction, quantile regression and hierarchical forecasting.
The listing recommends familiarity with Python and basic machine-learning concepts.

Price

$42.74Save 26%

Reference price: ~~$57.99~~

Price checked: Nov 14, 2026

View offer Ahead of Black FridayExplore the selection

Often bought together

Machine Learning for Algorithmic Trading: Predictive models to extract signals from market and alternative data for systematic trading strategies with Python

$47.60

Python for Algorithmic Trading Cookbook: Recipes for designing, building, and deploying algorithmic trading strategies with Python

$44.99

Pros and cons

Pros

Connects classical forecasting foundations with modern machine-learning and deep-learning approaches in one structured volume.
Includes practical workflow topics such as data preparation, feature engineering, evaluation and handling seasonality or non-stationarity.
Probabilistic forecasting and sparse or intermittent series broaden its relevance beyond point predictions.

Cons

The broad, advanced subject matter is likely a demanding fit for readers without Python or basic machine-learning familiarity.

Technical specifications

9 specifications

Book details
Publisher	Packt Publishing
Edition	2nd ed.
Publication date	October 31, 2024
Language	English
Print length	658 pages
ISBN-10	1835883184
ISBN-13	978-1835883181

Physical details
Weight listed	2.46 pounds
Dimensions	7.5 x 1.49 x 9.25 inches

What to know before buying

The listing describes this edition as adding deeper coverage of transformers, probabilistic forecasting, conformal prediction and hierarchical forecasting. Consider the listed 658-page length when choosing a desk reference or study text.

Who is it for?

A good fit for

Data scientists and machine-learning engineers working with time-dependent data.
Financial and quantitative analysts, researchers, and students with Python and basic machine-learning knowledge.
Readers seeking a progression from traditional forecasting methods into deep-learning and probabilistic approaches.

Less suited to

Readers seeking a short introductory overview or a book that does not assume Python and basic machine-learning familiarity.
Readers looking for a general Python or machine-learning reference rather than a time-series forecasting text.

Customer reviews on Amazon

4.3 / 5 · 46 ratings

5 stars75%75%
4 stars4%4%
3 stars7%7%
2 stars7%7%
1 star7%7%

Aggregate customer ratings lean positive, with most ratings in the five-star category, but the distribution also includes ratings across the lower categories. Without written feedback, the reasons for praise or criticism are unknown. Treat the ratings as a limited satisfaction signal rather than evidence about teaching quality or technical accuracy. The book's stated breadth makes checking its sample and topic coverage especially useful before choosing it.

Frequently asked questions

What background does the listing recommend?

It recommends familiarity with Python and basic machine-learning concepts, so readers without those foundations may find the advanced material challenging.

Which forecasting methods does the book cover?

The listing spans ARIMA and exponential smoothing through regression-based machine learning, global models, ensembles, neural networks and transformers.

Does it cover probabilistic forecasting?

Yes. The described topics include conformal prediction, Monte Carlo dropout, quantile regression and hierarchical forecasting.

How long is the listed edition?

The listing gives a print length of 658 pages.