Modern Time Series Analysis with R: Practical forecasting and impact estimation with tidy, reproducible workflows (Paperback or Softback)

Gain expertise in modern time series forecasting and causal inference in R to solve real-world business problems with reproducible, high-quality code

Key Features

• Explore forecasting and causal inference with practical R examples
• Build reproducible, high-quality time series workflows using tidyverse and modern R packages
• Apply models to real-world business scenarios with step-by-step guidance
• Purchase of the print or Kindle book includes a free PDF eBook

Book Description

Modern Time Series Analysis with R is a comprehensive, hands-on guide to mastering the art of time series analysis using the R programming language. Written by leading experts in applied statistics and econometrics, this book helps data scientists, analysts, and developers bridge the gap between traditional statistical theory and practical business applications.

Starting with the foundations of R and tidyverse, you’ll explore the core components of time series data, data wrangling, and visualization techniques. The chapters then guide you through key modeling approaches, ranging from classical methods like ARIMA and exponential smoothing to advanced computational techniques, such as machine learning, deep learning, and ensemble forecasting.

Beyond forecasting, you’ll discover how time series can be applied to causal inference, anomaly detection, change point analysis, and multiple time series modeling. Practical examples and reproducible code will empower you to assess business problems, choose optimal solutions, and communicate results effectively through dynamic R-based reporting.

By the end of this book, you’ll be confident in applying modern time series methods to real-world data, delivering actionable insights for strategic decision-making in business, finance, technology, and beyond.

What you will learn

• Understand the core concepts and structure of time series data
• Wrangle and visualize time series effectively
• Apply transformations and decomposition techniques
• Build and compare univariate forecasting models
• Apply statistical, ML, and DL models strategically based on context
• Forecast hierarchical and grouped time series
• Measure causal impact using interrupted time series analysis
• Detect anomalies, structural changes, and handle missing data

Who this book is for

This book is for data scientists, analysts, and developers who want to master time series analysis using R. It is ideal for professionals in finance, retail, technology, and research, as well as students seeking practical, business-oriented approaches to forecasting and causal inference. Basic knowledge of R is assumed, but no advanced mathematics is required.

Table of Contents

1. R, RStudio, and R packages
2. Objects and Functions in R
3. Data Input/Output in R
4. Time Series Characteristics
5. Time Series Data Wrangling and Visualization
6. Business Applications of Time Series Analysis
7. Time Series Adjustments, Transformations, and Decomposition
8. Time Series Features
9. Time Series Smoothing and Filtering
10. Basics of Forecasting
11. Exponential Smoothing
12. ARIMA Forecasting Models
13. Advanced Computational Methods for Forecasting
14. Forecasting Models for Multiple Time Series
15. Causal Impact Estimation
16. Changepoint Detection
17. Anomaly Detection and Imputation

Dr. Yeasmin Khandakar is a data scientist with over 15 years of experience across diverse sectors, including FinTech (Portland House Group), MedTech (Optalert), retail (Coles, Officeworks) and transport (Transurban). She has a PhD from Monash University and is the co-author of the paper Automatic time series forecasting: the forecast package for R, which has generated over 5,900+ citations. Dr. Khandakar specializes in solving strategic business challenges by integrating advanced statistical methods with machine learning and deep learning, and robust time-series techniques.

Dr. Roman Ahmed is an experienced statistician with a PhD specializing in time-series forecasting. He has more than two decades of experience across the corporate and academic sectors. With a career including prominent technical leadership at Optus, Xero, and ANZ Bank, he excels at applying high-impact forecasting, econometric, and machine learning solutions to business strategy. Roman has published methodological and applied research in top-tier journals and has presented work at prestigious conferences. His expertise lies in translating sophisticated methodological research into scalable, real-world tools, particularly within the R ecosystem.