book

Modern Time Series Analysis with R

Name: Modern Time Series Analysis with R
ISBN: 9781805124016

by Dr. Yeasmin Khandakar, Dr. Roman Ahmed

February 2026

Intermediate to advanced

628 pages

18h 6m

English

Packt Publishing

Read now

Unlock full access

Preface
Free benefits with your book
Part 1: Setting the Scene: R
Chapter 1: R, Rstudio, and R packages
1.1 Introduction to the R ecosystem1.1.1 What is R?1.1.2 What is RStudio?1.1.3 What are R packages?1.2 Downloading R and RStudio1.3 Installing R and RStudio with admin access1.3.1 Windows1.3.2 macOS1.3.3 Linux1.3.3.1 Ubuntu1.3.3.2 Debian1.4 Installation in a custom location1.5 Installing R packages1.5.1 Package installation via the RStudio menu1.5.2 Package installation via R console1.5.3 R package pathways1.5.4 Installing R packages in development1.6 Loading installed packages1.7 Updating software and packages1.7.1 Checking and updating R packages1.7.2 Checking and updating R and RStudio1.8 How to choose R packages1.9 RStudio for productive programming1.9.1 Do configure global and project level options1.9.2 Do set up a working directory1.9.3 Do start with an RStudio project1.9.4 Do not preserve the workspace between sessions1.10 The tidyverse package collectionSummaryFurther readingGet this book’s PDF version and more
Chapter 2: Objects and Functions in R
Technical requirements2.1 Objects and their names2.2 Object types2.3 Data types2.4 Atomic vectors2.4.1 Double2.4.2 Integer2.4.3 Logical2.4.4 Character2.5 Lists2.5.1 Matrix2.5.2 Array2.6 Coercion2.7 S3 atomic vectors2.7.1 Factors2.7.2 Dates2.7.3 Date-times2.7.4 Duration2.8 S3 lists2.8.1 Data frames2.8.2 Tibbles2.9 Missing values2.10 Time Series Specific Objects2.10.1 ts2.10.2 tsibble2.11 Functions in R2.12 Comparisons2.13 Conditions2.14 Conditional execution2.15 Iterations2.15.1 The apply family2.15.2 The map familySummaryFurther reading
Chapter 3: Data Input/Output in R
Technical requirements3.1 General structure of input/output functions3.2 Saving and loading files generated by R3.3 Data files from R packages3.4 Importing external files3.4.1 The readr package3.4.2 The readxl package3.4.3 The fst package3.4.4 Import data via RStudio’s menu3.5 Importing data from relational databases3.5.1 Connecting to databases3.5.2 Credential security3.5.2.1 Pre-configuring credentials during ODBC set-up3.5.2.2 The keyring package3.5.2.3 The config package3.5.2.4 The RStudio API3.6 Using SQL from R3.6.1 Inline SQL query3.6.2 SQL query from a file3.6.3 SQL translations via dplyr3.6.4 Parameterized query using glue_sql()3.7 Data governanceSummaryFurther readingGet this book’s PDF version and more
Part 2: The Main Character: Time Series
Chapter 4: Time Series Characteristics
4.1 What is a time series?4.2 Distinction of time series from other time-indexed data4.3 How to classify time-indexed data to apply suitable analytics4.4 Components of a time series4.4.1 Trend4.4.2 Seasonality4.4.3 Cyclical4.4.4 Random errors or remainders4.5 Combining time series components4.6 Stationarity4.7 Autocorrelation4.8 Types of time seriesSummary
Chapter 5: Time Series Data Wrangling and Visualization
Technical requirements5.1 Parsing date-time variables5.2 Setting components of date-time5.3 When to transform data into time series object?5.4 Wrangling functions5.5 Visualization functions5.6 Time plots5.7 Seasonal plots5.7.1 Seasonal subseries plots5.8 Visualizing relationships among time series5.8.1 Scatter plots5.8.2 Lag plots5.9 Interactive plotsSummaryGet this book’s PDF version and more
Chapter 6: Business Applications of Time Series Analysis
Technical requirements6.1 Time series characteristics level problem domains6.1.1 Trend (trend-cycle) analysis6.1.2 Seasonality analysis6.1.3 Outlier/anomaly detection6.2 Inference and attribution problem domains6.3 Impact measurement problem domains6.4 Forecasting problem domainsSummaryFurther reading
Part 3: The Makeover: Adjusting the Appearance

Chapter 7: Time Series Adjustments, Transformations, and Decomposition
Technical requirements7.1 Summary of differences: Adjustments, transformations and decompositions7.2 Rationales for time series adjustments and transformations7.3 Time series adjustments7.3.1 Calendar adjustments7.3.2 Population adjustments7.3.3 Base adjustments7.4 Time series transformations7.4.1 Logarithmic transformation7.4.2 Box-Cox transformation7.5 Decomposition plots7.6 Decomposition using MAs7.6.1 MA smoothing of data with trends7.6.2 MA smoothing of data with seasonality7.7 Trend and seasonal decomposition with the classical method7.8 Model based decomposition methods7.9 STL decomposition7.9.1 Summary of decomposition techniquesSummaryFurther readingGet this book’s PDF version and more
Chapter 8: Time Series Features
Technical requirements8.1 Time series features8.2 Statistical summary-based features8.3 Autocorrelation based features8.4 STL-based features8.5 Tiling window features8.6 Sliding window features8.7 Statistical test-based features8.8 Other useful features8.9 Clustering using time series features8.9.1 PCA on time series features8.9.2 Clustering using time series features, UMAP, and K-meansSummaryFurther reading
Chapter 9: Time Series Smoothing and Filtering
Technical requirements9.1 Comparison of time series preprocessing techniques9.2 Exponentially weighted moving average method (EWMA)9.3 Time series filtering9.3.1 The Nyquist frequency and sampling considerations9.4 Linear filtering with moving average (MA)9.5 The Hodrick-Prescott filter9.6 The Kalman filterSummaryFurther readingGet this book’s PDF version and more
Part 4: The Crystal Ball: Forecasting
Chapter 10: Basics of Forecasting
Technical requirements10.1 Notations10.2 Forecasting workflow10.3 fable for forecasting10.4 Naïve forecasting method10.5 Seasonal Naïve method10.6 Forecasting using average10.7 Time Series Linear Model (TSLM)10.7.1 Scenario analysis10.8 Evaluating point forecast accuracy10.9 Time series cross-validationSummaryFurther reading
Chapter 11: Exponential Smoothing
Technical requirements11.1 Forecasting methods versus models11.2 Exponential smoothing forecasting11.3 Simple Exponential Smoothing (SES)11.3.1 Component form of the (N, N) method11.3.2 SES with additive errors: ETS(A, N, N)11.3.3 SES with multiplicative errors: ETS(M, N, N)11.4 Holt’s linear trend method11.4.1 Component form of Holt’s linear trend method11.4.2 Holt’s linear trend model with additive errors: ETS(A, A, N)11.4.3 Holt’s linear trend model with multiplicative errors: ETS(M, A, N)11.4.4 Additive damped trend method11.5 Holt-Winters’ trend and seasonality method11.5.1 Component form of additive Holt-Winters’ method (A, A)11.5.2 Component form of multiplicative Holt-Winters’ method (A, M)11.5.3 Component form of Holt-Winters’ damped method (Ad, A)11.6 Automated model selection using ETS()SummaryFurther readingGet this book’s PDF version and more
Chapter 12: ARIMA Forecasting Models
Technical requirements12.1 Basics of ARIMA models12.2 Autoregressive model (the AR in ARIMA)12.3 Integration (the I in ARIMA)12.4 Moving Average model (the MA in ARIMA)12.5 General structure of an ARIMA model12.5.1 Non-seasonal ARIMA model12.5.2 Seasonal ARIMA models12.6 ARIMA with exogenous variables (ARIMAX)12.7 Steps in fitting an ARIMA model12.8 Automatic ARIMA algorithm12.9 ARIMA and forecast explainability12.9.1 Impact of the constant term on forecasts12.9.2 Prediction intervals from ARIMA models12.9.3 Impact of differencing orders on prediction intervals12.9.4 Impact of AR order on cyclic forecastsSummaryFurther reading
Chapter 13: Advanced Computational Methods for Forecasting
Technical requirements13.1 Understanding the connection between forecasting and prediction13.2 DL forecasting: Neural network13.2.1 Forecasting using neural networks13.2.2 Prediction interval for neural networks13.3 ML forecasting13.3.1 Application of ML in forecasting13.3.2 Quantile regularized regression: A brief summary13.4 Forecasting as a curve fitting: The Prophet method13.4.1 Prophet forecasting13.5 Forecast ensembleSummaryFurther readingGet this book’s PDF version and more
Chapter 14: Forecasting Models for Multiple Time Series
Technical requirements14.1 The dynamics of multiple interconnected time series14.2 Vector Autoregression model14.2.1 Mathematical background of the VAR(1) model14.2.2 Inspect time series for stationarity14.2.3 Identifying VAR model orders14.2.4 Interpretation of VAR model coefficients14.2.5 Granger causality test for a VAR(1) model14.2.6 Impulse Response Analysis14.2.7 Forecasting from a VAR model14.3 Hierarchical and grouped time series14.3.1 Traditional hierarchical forecasting: top-down, bottom-up, and middle-out approaches14.3.1.1 Top-down approach14.3.1.2 Bottom-up approach14.3.1.3 Middle-out approach14.3.2 Optimal reconciliation forecast for hierarchical time series using trace minimization14.3.2.1 Reconciliation problem Setup14.3.2.2 Optimal reconciliation via trace minimization14.3.3 Forecast accuracy evaluation14.4 Machine learning for multiple time series forecastingSummaryFurther reading
Part 5: Plot Twists and Turning Points: Measuring Impact and Detecting Anomalies
Chapter 15: Causal Impact Estimation
Technical requirements15.1 What is causal inference?15.2 Causal inference with experimentation15.2.1 Practical challenges of conducting experiments15.3 Causal inference with observational data15.4 Equivalence of hypothesis tests and regression15.4.1 Paired sample15.4.2 Two independent samples15.5 Interrupted Time Series (ITS) analysis15.6 ITS on stationary time series15.6.1 Stationary data impact measurement approaches15.6.2 Durbin-Watson test to identify an AR(1) process15.6.3 Statistical power analysis in regression15.7 ITS on time series with a trend15.7.1 Trend data impact measurement approaches15.8 ITS on time series with autocorrelation15.8.1 AR(1) data impact measurement approaches15.8.2 Adjustments in the linear model for AR(1) process15.8.3 ARIMAX for measuring the impact from an AR(1) process15.9 ITS on time series with AR(1) and trends15.9.1 ARIMAX for measuring the impact from AR(1) data with trends15.10 Flexibilities of ITS-based inference15.11 Monitoring guardrail metrics in ITSSummaryFurther readingGet this book’s PDF version and more
Chapter 16: Changepoint Detection
Technical requirements16.1 What is changepoint detection?16.2 Breaks for Additive Season and Trend16.2.1 BFAST implementation on non-seasonal data16.2.2 BFAST implementation on a seasonal series16.3 Pruned Exact Linear Time method16.3.1 PELT implementation on non-seasonal data16.3.2 PELT implementation on a seasonal series16.4 Bayesian changepoint detection16.4.1 BEAST implementation on non-seasonal series16.4.2 BEAST implementation on seasonal data16.5 Method selection guidelinesSummaryFurther reading
Chapter 17: Anomaly Detection and Imputation
Technical requirements17.1 Anomaly detection in time series17.2 STL decomposition-based anomaly detection17.3 Isolation forest-based anomaly detection17.4 Imputation in time series17.4.1 Linear interpolation-based imputation of anomalies17.4.2 Tidy workflow for simultaneous anomaly detection and imputation17.5 Guidelines for selecting an anomaly detection methodSummaryFurther readingGet this book’s PDF version and more
Chapter 18: Unlock Your Exclusive Benefits
Unlock this Book’s Free Benefits in 3 Easy Steps
Index

Content preview from Modern Time Series Analysis with R

Index

ARIMA models341, 342

for non-seasonal time series371 – 373

general structure356

non-seasonal ARIMA model357 – 361

seasonal ARIMA model361 – 367

ARIMA with exogenous variables (ARIMAX)368 – 371

Akaike’s Information Criterion (AIC)336, 389, 436

Anomaly detection

in time series561

Apollo Guidance Computer (AGC)251

Athena95

aliasing245

anomalies

linear interpolation-based imputation575, 576

anomaly detection159

method guidelines, selecting578, 579

apply() function69

array31, 40

as_tsibble() function59

atomic vector31, 32

character vector35, 36

double vector33

integer vector34

logical vector34, 35

attributes31

autocorrelation115, 116

autocorrelation function (ACF)115, 207

features209

automated model selection

with ETS() function333 – 336 ...

Become an O’Reilly member and get unlimited access to this title plus top books and audiobooks from O’Reilly and nearly 200 top publishers, thousands of courses curated by job role, 150+ live events each month,
and much more.

Read now

Unlock full access

More than 5,000 organizations count on O’Reilly

O’Reilly covers everything we've got, with content to help us build a world-class technology community, upgrade the capabilities and competencies of our teams, and improve overall team performance as well as their engagement.

Julian F.

Head of Cybersecurity

I wanted to learn C and C++, but it didn't click for me until I picked up an O'Reilly book. When I went on the O’Reilly platform, I was astonished to find all the books there, plus live events and sandboxes so you could play around with the technology.

Addison B.

Field Engineer

I’ve been on the O’Reilly platform for more than eight years. I use a couple of learning platforms, but I'm on O'Reilly more than anybody else. When you're there, you start learning. I'm never disappointed.

Amir M.

Data Platform Tech Lead

I'm always learning. So when I got on to O'Reilly, I was like a kid in a candy store. There are playlists. There are answers. There's on-demand training. It's worth its weight in gold, in terms of what it allows me to do.

Mark W.

Embedded Software Engineer

Multivariate Time Series Analysis and Applications

Publisher Resources

ISBN: 9781805124016

Cloud Computing

Data Engineering

Data Science

AI & ML

Programming Languages

Software Architecture

IT/Ops

Security

Design

Business

Soft Skills

Modern Time Series Analysis with R

by Dr. Yeasmin Khandakar, Dr. Roman Ahmed

Index

Become an O’Reilly member and get unlimited access to this title plus top books and audiobooks from O’Reilly and nearly 200 top publishers, thousands of courses curated by job role, 150+ live events each month,
and much more.