The R Primer

The R Primer provides a collection of concise examples with solutions and interpretations of R problems frequently encountered by new users of this statistical software. The R Primer contains numerous examples that illustrate a specific situation, topic, or problem, including data importing, data management, classical statistical analyses, and high-quality graphical production. Each example is self-contained and includes R code that can be run exactly as shown, enabling results from the book to be replicated.

New to the Second Edition:

• Completely revised and updated with suggestions for using new and improved R packages
• Expanded with over 100 more pages
• New solutions for covering areas from web scraping over data wrangling to waffle plots and hanging rootograms.
• Additional intermediate and advanced topics in statistical data analysis including non-parametric statistics, random forests, penalized regression and curve smoothing.

The site contains source code, extra material and errata.

Summary

• Presents commonly encountered problems and solutions in R in a “cookbook” style
• Covers importing and transforming data, making statistical analyses, and creating graphics
• Provides an ideal introduction for beginning users of R
• Includes code for all examples

“This book provides a good introduction to R, using a clear layout and detailed, reproducible examples. An ideal tool for any new R user. … A wide range of topics are covered, making the book suitable for a variety of readers, from undergraduate students to professionals new to R … an extremely helpful introduction to a very useful statistical package.”

• Claire Keeble, Journal of Applied Statistics, 2012

Table of contents

• Preface
• Importing and exporting data
  • Import an R dataset from a package
  • Load and save R data files
  • Read and write text files
  • Read and write CSV files
  • Read data from the clipboard
  • Spreadsheets
  • Read and write data from a spreadsheet
  • Read and write Excel files
  • Statistical software programs
  • Import and export SAS datasets
  • Import and export SPSS datasets
  • Import or export a Stata dataset
  • Import a Systat dataset
  • Data exchange formats and databases
  • Import a JSON dataset
  • Read data from a simple XML file
  • Read data from an XML file
  • Export a data frame to XML
  • Import data from an HTML table
  • Scrape data from an HTML web page
  • Import from a MySQL/PostgreSQL database
  • Read data from an SQL database using ODBC
• Manipulating data
  • Use mathematical functions
  • Use common vector operations
  • Work with character vectors
  • Read non-ASCII character vectors
  • Sort and order data
  • Transform a variable
  • Find the value of \(x\) corresponding to the maximum or minimum of \(y\)
  • Check if elements in one object are present in another object
  • Apply a function to subsets of a vector
  • Fill in missing values with previous values
  • Convert comma as decimal mark to period
  • Lag or shift a vector
  • Calculate the area under a curve
  • Matrices, arrays, and tables
  • Apply function to margins of a matrix/array
  • Compute a matrix/array of proportions
  • Transpose a matrix (or data frame)
  • Create a table of counts
  • Convert a table of counts to a data frame
  • Dates and times
  • Parsing dates and times
  • Extract and format date/time information
  • Factors
  • Convert a factor to numeric
  • Convert a factor to character strings
  • Add a new level to an existing factor
  • Combine the levels of a factor
  • Remove unused levels of a factor
  • Change the reference level
  • Cut a numeric vector into a factor
  • Data frames and lists
  • Select a subset of a data frame
  • Select the complete cases of a data frame
  • Delete a variable from a data frame
  • Apply function to each variable in a data frame
  • Split data frame into subsets and apply function to each part
  • Apply function to each row of a data frame
  • Combine two datasets
  • Merge datasets
  • Add new observations to a data frame
  • Stack the columns of a data frame together
  • Reshape a data frame from wide to long format or vice versa
  • Convert a data frame to a vector
• Statistical analyses
  • Descriptive statistics
  • Compute summary statistics
  • Create descriptive table
  • Linear models
  • Fit a linear regression model
  • Fit a multiple linear regression model
  • Fit a polynomial regression model
  • Fit a one-way analysis of variance
  • Make post-hoc pairwise comparisons
  • Fit a two-way analysis of variance
  • Fit a linear normal model
  • Fit a penalized regression model
  • Generalized linear models
  • Fit a logistic regression model
  • Fit a conditional logistic regression model
  • Fit an ordinal logistic regression model
  • Fit a multinomial logistic regression model
  • Fit a Poisson regression model
  • Methods for analysis of repeated measurements
  • Fit a linear mixed-effects model
  • Fit a linear mixed-effects model with serial correlation
  • Fit a generalized linear mixed model
  • Fit a generalized estimating equation model
  • Analyze time series using an ARIMA model
  • Decompose a time series into trend, seasonal, and residual components
  • Specific methods
  • Compare populations using \(t\) test
  • Fit a nonlinear model
  • Fit a censored regression model
  • Fit a zero-inflated regression model
  • Fit a smooth curve
  • Fit a generalized additive model
  • Use meta analysis to combine and summarize the results from several studies
  • Use random forest for classification and regression
  • Fit a linear quantile regression model
  • Model validation
  • Test for normality of a single sample
  • Test variance homogeneity across groups
  • Validate a linear or generalized linear model
  • Contingency tables
  • Analyze two-dimensional contingency tables
  • Analyze two-dimensional contingency tables with ordinal categories
  • Analyze two-dimensional contingency tables with paired measurements
  • Analyze contingency tables using log-linear models
  • Agreement
  • Create a Bland–Altman plot of agreement to compare two quantitative methods
  • Determine agreement among several methods of a quantitative measurement
  • Calculate Cohen’s kappa for agreement
  • Multivariate methods
  • Fit a multivariate regression model
  • Cluster observations
  • Use principal component analysis to reduce data dimensionality
  • Fit a principal component regression model
  • Classify+\thinmuskip {.1667em observations+\thinmuskip {.1667em using+\thinmuskip {.1667em linear+\thinmuskip {.1667em discriminant +\thinmuskip {.1667em analysis
  • Use partial least squares regression for prediction
  • Resampling statistics and bootstrapping
  • Non-parametric bootstrap analysis
  • Use cross-validation to estimate the performance of a model or algorithm
  • Calculate power or sample size for simple designs
  • Non-parametric methods
  • Test median with Wilcoxon’s signed rank test
  • Use Mann–Whitney’s test to compare two groups
  • Compare groups using Kruskal–Wallis’ test
  • Compare groups using Friedman’s test for a two-way design
  • Survival analysis
  • Fit a Kaplan–Meier survival curve to event history data
  • Fit a Cox regression model (proportional hazards model)
  • Fit a Cox regression model (proportional hazards model) with time-varying covariates
  • Miscellaneous
  • Correct \(p\)-values for multiple testing
  • Use a Box–Cox transformation to make non-normally distributed data approximately normal
• Graphics
  • Use Greek letters and formulas in graphs
  • Set colors in R graphics
  • Set color palettes in R graphics
  • High-level plots
  • Create a scatter plot
  • Create a bubble plot
  • Create a histogram
  • Create a hanging rootogram
  • Create a violin- or boxplot
  • Create a bar plot
  • Create a bar plot with error bars
  • Create a plot with estimates and confidence intervals
  • Create a pie chart
  • Create a pyramid plot
  • Plot multiple series
  • Make a 2D surface plot
  • Make a 3D surface plot
  • Plot a 3D scatter plot
  • Create a heat map plot
  • Plot a correlation matrix
  • Make a quantile-quantile plot
  • Graphical model validation for linear models
  • Create a Venn or Euler diagram
  • Tweaking graphics
  • Add a broken axis to indicate discontinuity
  • Create a plot with two \(y\)-axes
  • Rotate axis labels
  • Multiple plots
  • Add a legend to a plot
  • Add a table to a plot
  • Label points in a scatter plot
  • Identify points in a scatter plot
  • Visualize points, shapes, and surfaces in 3D and interact with them in real-time
  • Working with graphics
  • Exporting graphics
  • Produce graphics in LaTeX-ready format
  • Embed fonts in postscript or pdf graphics
• R
  • Getting information
  • Getting help
  • Finding R source code for a function
  • Timing R commands and functions
  • R packages
  • Installing R packages
  • Update installed R packages
  • Unload a package
  • Install an R package from a repository
  • Install a package from BioConductor
  • List the installed packages
  • List the content of a package
  • Uninstall an R package
  • List or view vignettes
  • Permanently change the default directory where R installs packages
  • Automatically load a package at startup
  • The R workspace
  • List or delete objects
  • Change the current working directory
  • Save and load workspaces
  • Save and restore histories
  • Interact with the file system
  • Locate and choose files interactively
  • Interact with the operating system
  • Get session information
• Bibliography
• Index

About the author

Claus Thorn Ekstrøm is Professor at the Section of Biostatistics, University of Copenhagen where he teaches courses on statistics and R for beginners and advanced users. Professor Ekstrøm’s primary research interests lie within statistical genetics, genetic epidemiology, and bioinformatics, in particular genetic association studies, image analysis of microarray scans, and integrated analysis of gene expression and metabolic profile data.

Differences between the 1st and 2nd editions

The 2nd edition has updated the existing solutions to reflect new packages and improvements in R and an extra 33% material has been added.

---

Created by Claus Ekstrøm 2017, page last updated 24 October, 2020