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 core concepts and components of time series data
- Wrangle and visualize time series with tidyverse and R packages
- Apply ARIMA, exponential smoothing, and machine learning methods
- Explore deep learning and ensemble forecasting approaches
- Conduct causal inference with interrupted time series analysis
- Detect anomalies, structural changes, and perform change point analysis
- Analyze multiple time series using hierarchical and grouped models
- Automate reproducible reporting with RStudio and dynamic documents
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
- R, RStudio, and R packages
- Objects and Functions in R
- Data Input/Output in R
- Time Series Characteristics
- Time Series Data Wrangling and Visualization
- Business Applications of Time Series Analysis
- Time Series Adjustments, Transformations, and Decomposition
- Time Series Features
- Time Series Smoothing and Filtering
- Basics of Forecasting
- Exponential Smoothing
- ARIMA Forecasting Models
- Advanced Computational Methods for Forecasting
- Forecasting Models for Multiple Time Series
- Causal Impact Estimation
- Changepoint Detection
- Anomaly Detection and Imputation
Ahmed, Roman: - Roman Ahmed, PhD is a data science and analytics leader with 15+ years across industry and academia, specializing in statistics, econometrics, forecasting, A B testing, and ML AI. He currently serves as an Analytics Manager (Senior Specialist) at Optus and has held senior roles at Jobs and Skills Australia, Xero, the Australian Taxation Office, ANZ, Telstra, and Experian. Roman has led teams to build experimentation platforms, benefit-measurement frameworks, and large-scale time-series and econometric models that inform executive decision-making. He works fluently across R, Python, SQL, and STATA, deploys solutions in cloud environments such as Microsoft Azure, and has applied NLP and clustering to unlock insights from unstructured data. Roman earned his PhD in Econometrics and Business Statistics from Monash University and a BSc in Applied Statistics from the University of Dhaka. He is NV1 cleared and bilingual (English, Bengali). He brings a rigorous, business-first perspective to Modern Time Series Analysis with R.
Khandakar, Yeasmin: - Yeasmin Khandakar, PhD is a data scientist and statistician with 10+ years of experience building time series and machine learning solutions across industry and academia. She developed an automatic ARIMA modeling framework during her doctoral research under Prof. Rob Hyndman and has published highly cited work on automatic ARIMA forecasting. In industry, Yeasmin has led analytics and forecasting initiatives as Senior Data Scientist at Transurban, Officeworks, Coles, and Optalert, and previously designed automated trading algorithms as Senior Research Scientist at Portland House Group. Her expertise spans time series filtering and forecasting, predictive modeling, clustering, and large-scale model development in R and Python. Yeasmin holds a PhD in Statistics from Monash University, is NV1 cleared, and is certified as a Professional Scrum Product Owner (PSPO I). She brings deep practical and research insight to Modern Time Series Analysis with R
