finelybook
Hands-On Time Series Analysis with R: Build accurate forecasting models using ARIMA, Prophet, TBATS, Bayesian methods, and deep learning 2nd Edition
Author(s): Diogo Alves de Resende (Author), Shuen Mei (Author)
- Publisher Finelybook 出版社: Packt Publishing
- Publication Date 出版日期: September 30, 2026
- Edition 版本: 2nd
- Language 语言: English
- Print length 页数: 402 pages
- ISBN-10: 180323878X
- ISBN-13: 9781803238784
Book Description
Master modern time series forecasting in R using real datasets, applied models like ARIMA and Prophet, and advanced tools like LSTM for business-ready analysis
Key Features
- Build a practical skills in time series forecasting using R from scratch
- Levereage traditional and deep learning models including ARIMA, Prophet, and LSTM
- Analyze and visualize real-world datasets to drive decisions and discover insights
- Purchase of the print or Kindle book includes a free PDF eBook
Book Description
Hands-on Time Series Analysis with R, Second Edition, is a comprehensive, practical guide to understanding and applying time series analysis using R. Designed for professionals and students alike, the book will help you dissect time-ordered data, uncover trends, detect anomalies, and forecast future events.
You’ll start with R fundamentals and data structures, progressing through time series visualization and decomposition,before covering classical forecasting methods like exponential smoothing and ARIMA. You’ll also familiarize yourself with cutting-edge forecasting techniques such as Bayesian modeling, TBATS, and multivariate forecasting.
Building upon the extraordinary success of the first edition, this new edition explores the powerful Prophet library, neural networks, and LSTM models to handle complex forecasting challenges. Unlike other similar books, it uses real-world case studies from domains like e-commerce and supply chain make the content highly applicable, while self-assessment questions and step-by-step tutorials reinforce key concepts.
By the end of this book, you’ll be able to confidently apply a wide array of forecasting methods to solve real-world business and research problems.
What you will learn
- Work with time series data to uncover underlying patterns and trends
- Implement forecasting models like ARIMA and Prophet to predict future values
- Create powerful data visualizations to communicate time series insights
- Leverage R for time-based analysis across various domains
- Build and fine-tune models for different data scenarios
- Detect anomalies and adjust models for improved accuracy
- Use deep learning techniques like LSTM for complex time series tasks
Who this book is for
This book is for junior to mid-level professionals in business intelligence, finance, and supply chain roles, as well as academics and students interested in data science. No prior expertise in time series analysis is required, but a working knowledge of basic R programming is helpful.
Table of Contents
- Getting Started with the Basics of R
- Understanding Time Series Data
- Working with Dates and Times
- Visualizing Time Series Data
- Exploratory Data Analysis for Time Series
- Exponential Smoothing and Holt-Winters
- The ARIMA Model Family
- Forecasting with TBATS Modeling
- Bayesian Approaches in Time Series Analysis
- Multi-Series Forecasting Techniques
- Prophet for Time Series
- Introducing Neural Networks and Time Series
- Deep Learning with LSTM Models
Editorial Reviews
Editorial Reviews
Review
“A model that merely fits the past is not yet a forecast you can trust, as new data will eventually judge it. Resende and Mei take that seriously: a model here isn’t finished when it fits, only when it has survived cross-validation and honest tuning. From Holt-Winters through ARIMA, TBATS, and Bayesian methods to Prophet and LSTM, each technique arrives with a real business problem attached. A practical guide for anyone who must defend a number.”
Greg Rafferty, author of Forecasting Time Series Data with Prophet and The Strategic Data Scientist
“The second edition of the hands-on guide provides a framework for time series analysis and forecasting modeling using R. Readers learn to manipulate time-dependent data structures and implement foundational forecasting models, expanding into advanced machine learning and deep learning algorithms. Practical, worked-out examples demonstrate how to leverage various dedicated R libraries to handle complex datasets. The case studies illustrate real-world applications, showing how to apply those models in real-world situations while addressing operational challenges. Complete with source code and interactive exercises, it serves as a robust reference for professionals at any level seeking to implement data-driven forecasting workflows in a business context.”
Dr. Yeasmin Khandakar, co-creator of the Hyndman-Khandakar algorithm for automatic ARIMA forecasting and co-author of Modern Time Series Analysis with R
“I have worked extensively with time-series analysis in economic modeling and finance, and I found this to be an excellent book. It guides readers clearly through both the underlying methodology and its practical implementation in R.
What distinguishes this book is its practitioner-focused approach, something that is often missing from other titles in the field. The ready-to-run code makes the material immediately useful, without unnecessary detours or wasted time.
I recommend adding this book to your library if you are looking for a practical, accessible guide to time-series analysis in R.”
Siamak Amirghodsi, Quantitative Investor, Architect, Author, Ex-Amazon & Ex-Google
About the Author
Diogo Alves de Resende is the founder of The Data Hero and Bitte, with deep expertise in time series and commercial analytics. Formerly a senior commercial analyst at Zalando, he has implemented forecasting models for demand planning and return rates. As a popular Udemy instructor with over 20,000 students, Diogo brings his educational background and entrepreneurial experience into accessible, real-world teaching. He holds an MSc in Management and continues to blend technical insight with applied business strategy.
Shuen Mei is a big data analytic platforms expert with 15+ years of experience in designing, building, and executing large-scale, enterprise-distributed financial systems with mission-critical low-latency requirements. He is certified in the Apache Spark, Cloudera Big Data platform, including Developer, Admin, and HBase. He is also a certified AWS solutions architect with emphasis on peta-byte range real-time data platform systems.
