Package 'modelsummary'

Description

A convenience function which can be passed to the coef_rename argument of the modelsummary function.

Usage

coef_rename(
  x,
  factor = TRUE,
  factor_name = TRUE,
  poly = TRUE,
  backticks = TRUE,
  titlecase = TRUE,
  underscore = TRUE,
  asis = TRUE
)

Arguments

Examples

library(modelsummary)
dat <- mtcars
dat$horse_power <- dat$hp
mod <- lm(mpg ~ horse_power + factor(cyl), dat)
modelsummary(mod, coef_rename = coef_rename)

Description

Persistent user settings for the modelsummary package

Usage

config_modelsummary(
  factory_default,
  factory_latex,
  factory_html,
  factory_markdown,
  startup_message,
  reset = FALSE
)

Arguments

Description

datasummary can use any summary function which produces one numeric or character value per variable. The examples section of this documentation shows how to define custom summary functions.

modelsummary also supplies several shortcut summary functions which can be used in datasummary() formulas: Min, Max, Mean, Median, Var, SD, NPercent, NUnique, Ncol, P0, P25, P50, P75, P100.

See the Details and Examples sections below, and the vignettes on the modelsummary website:

• https://modelsummary.com/

• https://modelsummary.com/vignettes/datasummary.html

Usage

datasummary(
  formula,
  data,
  output = getOption("modelsummary_output", default = "default"),
  fmt = 2,
  title = getOption("modelsummary_title", default = NULL),
  notes = getOption("modelsummary_notes", default = NULL),
  align = getOption("modelsummary_align", default = NULL),
  add_columns = getOption("modelsummary_add_columns", default = NULL),
  add_rows = getOption("modelsummary_add_rows", default = NULL),
  sparse_header = getOption("modelsummary_sparse_header", default = TRUE),
  escape = getOption("modelsummary_escape", default = TRUE),
  ...
)

Arguments

Details

Visit the 'modelsummary' website for more usage examples: https://modelsummary.com

The 'datasummary' function is a thin wrapper around the 'tabular' function from the 'tables' package. More details about table-making formulas can be found in the 'tables' package documentation: ?tables::tabular

Hierarchical or "nested" column labels are only available for these output formats: tinytable, kableExtra, gt, html, rtf, and LaTeX. When saving tables to other formats, nested labels will be combined to a "flat" header.

Version 2.0.0, kableExtra, and tinytable

Since version 2.0.0, modelsummary uses tinytable as its default table-drawing backend. Learn more at: https://vincentarelbundock.github.io/tinytable/",

Revert to kableExtra for one session:

options(modelsummary_factory_default = 'kableExtra') options(modelsummary_factory_latex = 'kableExtra') options(modelsummary_factory_html = 'kableExtra')

Global Options

The behavior of modelsummary can be modified by setting global options. In particular, most of the arguments for most of the package's functions cna be set using global options. For example:

• options(modelsummary_output = "modelsummary_list")

• options(modelsummary_statistic = '({conf.low}, {conf.high})')

• options(modelsummary_stars = TRUE)

Options not specific to given arguments are listed below.

Balance tables: summary functions

You can customize the summary statistics shown in datasummary_balance():

• options(modelsummary_balance_fn = list(Mean = Mean, "Std. Dev." = SD))

The list must be named, and each function should accept a numeric vector and return a single value. When this option is set and a weights column is present, weights are ignored with a warning.

Model labels: default column names

These global option changes the style of the default column headers:

• options(modelsummary_model_labels = "roman")

• options(modelsummary_model_labels_term = "Parameter")

• options(modelsummary_model_labels_group = "Group")

• options(modelsummary_model_labels_model = "Model")

The options above control the label shown in the stub columns for terms, groups, and model names. By default, those headers are blank. Setting any of these options to a character string inserts that label as the column header.

The supported styles are: "model", "arabic", "letters", "roman", "(arabic)", "(letters)", "(roman)"

Table-making packages

modelsummary supports 6 table-making packages: tinytable, kableExtra, gt, flextable, huxtable, and DT. Some of these packages have overlapping functionalities. To change the default backend used for a specific file format, you can use ' the options function:

options(modelsummary_factory_html = 'kableExtra') options(modelsummary_factory_word = 'huxtable') options(modelsummary_factory_png = 'gt') options(modelsummary_factory_latex = 'gt') options(modelsummary_factory_latex_tabular = 'kableExtra')

Table themes

Change the look of tables in an automated and replicable way, using the modelsummary theming functionality. See the vignette: https://modelsummary.com/vignettes/appearance.html

• modelsummary_theme_gt

• modelsummary_theme_kableExtra

• modelsummary_theme_huxtable

• modelsummary_theme_flextable

• modelsummary_theme_dataframe

Model extraction functions

modelsummary can use two sets of packages to extract information from statistical models: the easystats family (performance and parameters) and broom. By default, it uses easystats first and then falls back on broom in case of failure. You can change the order of priorities or include goodness-of-fit extracted by both packages by setting:

options(modelsummary_get = "easystats")

options(modelsummary_get = "broom")

options(modelsummary_get = "all")

The "all" option (default) means easystats then broom.

Formatting numeric entries

By default, LaTeX tables enclose all numeric entries in the ⁠\num{}⁠ command from the siunitx package. To prevent this behavior, or to enclose numbers in dollar signs (for LaTeX math mode), users can call:

options(modelsummary_format_numeric_latex = "plain")

options(modelsummary_format_numeric_latex = "mathmode")

A similar option can be used to display numerical entries using MathJax in HTML tables:

options(modelsummary_format_numeric_html = "mathjax")

LaTeX preamble

When creating LaTeX via the tinytable backend (default in version 2.0.0 and later), it is useful to include the following commands in the LaTeX preamble of your documents. These commands are automatically added to the preamble when compiling Rmarkdown or Quarto documents, except when the modelsummary() calls are cached.

\usepackage{tabularray}
\usepackage{float}
\usepackage{graphicx}
\usepackage[normalem]{ulem}
\UseTblrLibrary{booktabs}
\UseTblrLibrary{siunitx}
\newcommand{\tinytableTabularrayUnderline}[1]{\underline{#1}}
\newcommand{\tinytableTabularrayStrikeout}[1]{\sout{#1}}
\NewTableCommand{\tinytableDefineColor}[3]{\definecolor{#1}{#2}{#3}}

Examples

library(modelsummary)

# The left-hand side of the formula describes rows, and the right-hand side
# describes columns. This table uses the "mpg" variable as a row and the "mean"
# function as a column:

datasummary(mpg ~ mean, data = mtcars)

# This table uses the "mean" function as a row and the "mpg" variable as a column:

datasummary(mean ~ mpg, data = mtcars)

# Display several variables or functions of the data using the "+"
# concatenation operator. This table has 2 rows and 2 columns:

datasummary(hp + mpg ~ mean + sd, data = mtcars)

# Nest variables or statistics inside a "factor" variable using the "*" nesting
# operator. This table shows the mean of "hp" and "mpg" for each value of
# "cyl":

mtcars$cyl <- as.factor(mtcars$cyl)
datasummary(hp + mpg ~ cyl * mean, data = mtcars)

# If you don't want to convert your original data
# to factors, you can use the 'Factor()'
# function inside 'datasummary' to obtain an identical result:

datasummary(hp + mpg ~ Factor(cyl) * mean, data = mtcars)

# You can nest several variables or statistics inside a factor by using
# parentheses. This table shows the mean and the standard deviation for each
# subset of "cyl":

datasummary(hp + mpg ~ cyl * (mean + sd), data = mtcars)

# Summarize all numeric variables with 'All()'
datasummary(All(mtcars) ~ mean + sd, data = mtcars)

# Define custom summary statistics. Your custom function should accept a vector
# of numeric values and return a single numeric or string value:

minmax <- function(x) sprintf("[%.2f, %.2f]", min(x), max(x))
mean_na <- function(x) mean(x, na.rm = TRUE)

datasummary(hp + mpg ~ minmax + mean_na, data = mtcars)

# To handle missing values, you can pass arguments to your functions using
# '*Arguments()'

datasummary(hp + mpg ~ mean * Arguments(na.rm = TRUE), data = mtcars)

# For convenience, 'modelsummary' supplies several convenience functions
# with the argument `na.rm=TRUE` by default: Mean, Median, Min, Max, SD, Var,
# P0, P25, P50, P75, P100, NUnique, Histogram

#datasummary(hp + mpg ~ Mean + SD + Histogram, data = mtcars)

# These functions also accept a 'fmt' argument which allows you to
# round/format the results

datasummary(hp + mpg ~ Mean * Arguments(fmt = "%.3f") + SD * Arguments(fmt = "%.1f"), data = mtcars)

# Save your tables to a variety of output formats:
f <- hp + mpg ~ Mean + SD
#datasummary(f, data = mtcars, output = 'table.html')
#datasummary(f, data = mtcars, output = 'table.tex')
#datasummary(f, data = mtcars, output = 'table.md')
#datasummary(f, data = mtcars, output = 'table.docx')
#datasummary(f, data = mtcars, output = 'table.pptx')
#datasummary(f, data = mtcars, output = 'table.jpg')
#datasummary(f, data = mtcars, output = 'table.png')

# Display human-readable code
#datasummary(f, data = mtcars, output = 'html')
#datasummary(f, data = mtcars, output = 'markdown')
#datasummary(f, data = mtcars, output = 'latex')

# Return a table object to customize using a table-making package
#datasummary(f, data = mtcars, output = 'tinytable')
#datasummary(f, data = mtcars, output = 'gt')
#datasummary(f, data = mtcars, output = 'kableExtra')
#datasummary(f, data = mtcars, output = 'flextable')
#datasummary(f, data = mtcars, output = 'huxtable')

# add_rows
new_rows <- data.frame(a = 1:2, b = 2:3, c = 4:5)
attr(new_rows, 'position') <- c(1, 3)
datasummary(mpg + hp ~ mean + sd, data = mtcars, add_rows = new_rows)

References

Arel-Bundock V (2022). “modelsummary: Data and Model Summaries in R.” Journal of Statistical Software, 103(1), 1-23. doi:10.18637/jss.v103.i01.'

Description

Creates balance tables with summary statistics for different subsets of the data (e.g., control and treatment groups). It can also be used to create summary tables for full data sets. See the Details and Examples sections below, and the vignettes on the modelsummary website:

• https://modelsummary.com/

• https://modelsummary.com/vignettes/datasummary.html

Usage

datasummary_balance(
  formula,
  data,
  output = getOption("modelsummary_output", default = "default"),
  fmt = fmt_decimal(digits = 1, pdigits = 3),
  title = getOption("modelsummary_title", default = NULL),
  notes = getOption("modelsummary_notes", default = NULL),
  align = getOption("modelsummary_align", default = NULL),
  stars = getOption("modelsummary_stars", default = FALSE),
  add_columns = getOption("modelsummary_add_columns", default = NULL),
  add_rows = getOption("modelsummary_add_rows", default = NULL),
  dinm = getOption("modelsummary_dinm", default = TRUE),
  dinm_statistic = getOption("modelsummary_dinm_statistic", default = "std.error"),
  fn = getOption("modelsummary_balance_fn", default = NULL),
  escape = getOption("modelsummary_escape", default = TRUE),
  ...
)

Arguments

Version 2.0.0, kableExtra, and tinytable

Since version 2.0.0, modelsummary uses tinytable as its default table-drawing backend. Learn more at: https://vincentarelbundock.github.io/tinytable/",

Revert to kableExtra for one session:

options(modelsummary_factory_default = 'kableExtra') options(modelsummary_factory_latex = 'kableExtra') options(modelsummary_factory_html = 'kableExtra')

Global Options

The behavior of modelsummary can be modified by setting global options. In particular, most of the arguments for most of the package's functions cna be set using global options. For example:

• options(modelsummary_output = "modelsummary_list")

• options(modelsummary_statistic = '({conf.low}, {conf.high})')

• options(modelsummary_stars = TRUE)

Options not specific to given arguments are listed below.

Balance tables: summary functions

You can customize the summary statistics shown in datasummary_balance():

• options(modelsummary_balance_fn = list(Mean = Mean, "Std. Dev." = SD))

The list must be named, and each function should accept a numeric vector and return a single value. When this option is set and a weights column is present, weights are ignored with a warning.

Model labels: default column names

These global option changes the style of the default column headers:

• options(modelsummary_model_labels = "roman")

• options(modelsummary_model_labels_term = "Parameter")

• options(modelsummary_model_labels_group = "Group")

• options(modelsummary_model_labels_model = "Model")

The options above control the label shown in the stub columns for terms, groups, and model names. By default, those headers are blank. Setting any of these options to a character string inserts that label as the column header.

The supported styles are: "model", "arabic", "letters", "roman", "(arabic)", "(letters)", "(roman)"

Table-making packages

modelsummary supports 6 table-making packages: tinytable, kableExtra, gt, flextable, huxtable, and DT. Some of these packages have overlapping functionalities. To change the default backend used for a specific file format, you can use ' the options function:

options(modelsummary_factory_html = 'kableExtra') options(modelsummary_factory_word = 'huxtable') options(modelsummary_factory_png = 'gt') options(modelsummary_factory_latex = 'gt') options(modelsummary_factory_latex_tabular = 'kableExtra')

Table themes

Change the look of tables in an automated and replicable way, using the modelsummary theming functionality. See the vignette: https://modelsummary.com/vignettes/appearance.html

• modelsummary_theme_gt

• modelsummary_theme_kableExtra

• modelsummary_theme_huxtable

• modelsummary_theme_flextable

• modelsummary_theme_dataframe

Model extraction functions

modelsummary can use two sets of packages to extract information from statistical models: the easystats family (performance and parameters) and broom. By default, it uses easystats first and then falls back on broom in case of failure. You can change the order of priorities or include goodness-of-fit extracted by both packages by setting:

options(modelsummary_get = "easystats")

options(modelsummary_get = "broom")

options(modelsummary_get = "all")

The "all" option (default) means easystats then broom.

Formatting numeric entries

By default, LaTeX tables enclose all numeric entries in the ⁠\num{}⁠ command from the siunitx package. To prevent this behavior, or to enclose numbers in dollar signs (for LaTeX math mode), users can call:

options(modelsummary_format_numeric_latex = "plain")

options(modelsummary_format_numeric_latex = "mathmode")

A similar option can be used to display numerical entries using MathJax in HTML tables:

options(modelsummary_format_numeric_html = "mathjax")

LaTeX preamble

When creating LaTeX via the tinytable backend (default in version 2.0.0 and later), it is useful to include the following commands in the LaTeX preamble of your documents. These commands are automatically added to the preamble when compiling Rmarkdown or Quarto documents, except when the modelsummary() calls are cached.

\usepackage{tabularray}
\usepackage{float}
\usepackage{graphicx}
\usepackage[normalem]{ulem}
\UseTblrLibrary{booktabs}
\UseTblrLibrary{siunitx}
\newcommand{\tinytableTabularrayUnderline}[1]{\underline{#1}}
\newcommand{\tinytableTabularrayStrikeout}[1]{\sout{#1}}
\NewTableCommand{\tinytableDefineColor}[3]{\definecolor{#1}{#2}{#3}}

Examples

library(modelsummary)
datasummary_balance(~am, mtcars)

# custom summary functions
fun <- list(
  "Median" = function(x) median(x, na.rm = TRUE),
  "Variance" = function(x) var(x, na.rm = TRUE)
)
datasummary_balance(~am, mtcars, fn = fun)

References

Arel-Bundock V (2022). “modelsummary: Data and Model Summaries in R.” Journal of Statistical Software, 103(1), 1-23. doi:10.18637/jss.v103.i01.'

Description

The names of the variables displayed in the correlation table are the names of the columns in the data. You can rename those columns (with or without spaces) to produce a table of human-readable variables. See the Details and Examples sections below, and the vignettes on the modelsummary website:

• https://modelsummary.com/

• https://modelsummary.com/vignettes/datasummary.html

Usage

datasummary_correlation(
  data,
  output = getOption("modelsummary_output", default = "default"),
  method = getOption("modelsummary_method", default = "pearson"),
  fmt = 2,
  align = getOption("modelsummary_align", default = NULL),
  add_rows = getOption("modelsummary_add_rows", default = NULL),
  add_columns = getOption("modelsummary_add_columns", default = NULL),
  title = getOption("modelsummary_title", default = NULL),
  notes = getOption("modelsummary_notes", default = NULL),
  escape = getOption("modelsummary_escape", default = TRUE),
  stars = getOption("modelsummary_stars", default = FALSE),
  ...
)

Arguments

Version 2.0.0, kableExtra, and tinytable

Since version 2.0.0, modelsummary uses tinytable as its default table-drawing backend. Learn more at: https://vincentarelbundock.github.io/tinytable/",

Revert to kableExtra for one session:

options(modelsummary_factory_default = 'kableExtra') options(modelsummary_factory_latex = 'kableExtra') options(modelsummary_factory_html = 'kableExtra')

Global Options

The behavior of modelsummary can be modified by setting global options. In particular, most of the arguments for most of the package's functions cna be set using global options. For example:

• options(modelsummary_output = "modelsummary_list")

• options(modelsummary_statistic = '({conf.low}, {conf.high})')

• options(modelsummary_stars = TRUE)

Options not specific to given arguments are listed below.

Balance tables: summary functions

You can customize the summary statistics shown in datasummary_balance():

• options(modelsummary_balance_fn = list(Mean = Mean, "Std. Dev." = SD))

The list must be named, and each function should accept a numeric vector and return a single value. When this option is set and a weights column is present, weights are ignored with a warning.

Model labels: default column names

These global option changes the style of the default column headers:

• options(modelsummary_model_labels = "roman")

• options(modelsummary_model_labels_term = "Parameter")

• options(modelsummary_model_labels_group = "Group")

• options(modelsummary_model_labels_model = "Model")

The options above control the label shown in the stub columns for terms, groups, and model names. By default, those headers are blank. Setting any of these options to a character string inserts that label as the column header.

The supported styles are: "model", "arabic", "letters", "roman", "(arabic)", "(letters)", "(roman)"

Table-making packages

modelsummary supports 6 table-making packages: tinytable, kableExtra, gt, flextable, huxtable, and DT. Some of these packages have overlapping functionalities. To change the default backend used for a specific file format, you can use ' the options function:

options(modelsummary_factory_html = 'kableExtra') options(modelsummary_factory_word = 'huxtable') options(modelsummary_factory_png = 'gt') options(modelsummary_factory_latex = 'gt') options(modelsummary_factory_latex_tabular = 'kableExtra')

Table themes

Change the look of tables in an automated and replicable way, using the modelsummary theming functionality. See the vignette: https://modelsummary.com/vignettes/appearance.html

• modelsummary_theme_gt

• modelsummary_theme_kableExtra

• modelsummary_theme_huxtable

• modelsummary_theme_flextable

• modelsummary_theme_dataframe

Model extraction functions

modelsummary can use two sets of packages to extract information from statistical models: the easystats family (performance and parameters) and broom. By default, it uses easystats first and then falls back on broom in case of failure. You can change the order of priorities or include goodness-of-fit extracted by both packages by setting:

options(modelsummary_get = "easystats")

options(modelsummary_get = "broom")

options(modelsummary_get = "all")

The "all" option (default) means easystats then broom.

Formatting numeric entries

By default, LaTeX tables enclose all numeric entries in the ⁠\num{}⁠ command from the siunitx package. To prevent this behavior, or to enclose numbers in dollar signs (for LaTeX math mode), users can call:

options(modelsummary_format_numeric_latex = "plain")

options(modelsummary_format_numeric_latex = "mathmode")

A similar option can be used to display numerical entries using MathJax in HTML tables:

options(modelsummary_format_numeric_html = "mathjax")

LaTeX preamble

When creating LaTeX via the tinytable backend (default in version 2.0.0 and later), it is useful to include the following commands in the LaTeX preamble of your documents. These commands are automatically added to the preamble when compiling Rmarkdown or Quarto documents, except when the modelsummary() calls are cached.

\usepackage{tabularray}
\usepackage{float}
\usepackage{graphicx}
\usepackage[normalem]{ulem}
\UseTblrLibrary{booktabs}
\UseTblrLibrary{siunitx}
\newcommand{\tinytableTabularrayUnderline}[1]{\underline{#1}}
\newcommand{\tinytableTabularrayStrikeout}[1]{\sout{#1}}
\NewTableCommand{\tinytableDefineColor}[3]{\definecolor{#1}{#2}{#3}}

Examples

library(modelsummary)

# clean variable names (base R)
dat <- mtcars[, c("mpg", "hp")]
colnames(dat) <- c("Miles / Gallon", "Horse Power")
datasummary_correlation(dat)

# clean variable names (tidyverse)
library(tidyverse)
dat <- mtcars %>%
  select(`Miles / Gallon` = mpg,
         `Horse Power` = hp)
datasummary_correlation(dat)

# `correlation` package objects
if (requireNamespace("correlation", quietly = TRUE)) {
  co <- correlation::correlation(mtcars[, 1:4])
  datasummary_correlation(co)

  # add stars to easycorrelation objects
  datasummary_correlation(co, stars = TRUE)
}

# alternative methods
datasummary_correlation(dat, method = "pearspear")

# custom function
cor_fun <- function(x) cor(x, method = "kendall")
datasummary_correlation(dat, method = cor_fun)

# rename columns alphabetically and include a footnote for reference
note <- sprintf("(%s) %s", letters[1:ncol(dat)], colnames(dat))
note <- paste(note, collapse = "; ")

colnames(dat) <- sprintf("(%s)", letters[1:ncol(dat)])

datasummary_correlation(dat, notes = note)

# `datasummary_correlation_format`: custom function with formatting
dat <- mtcars[, c("mpg", "hp", "disp")]

cor_fun <- function(x) {
  out <- cor(x, method = "kendall")
  datasummary_correlation_format(
    out,
    fmt = 2,
    upper_triangle = "x",
    diagonal = ".")
}

datasummary_correlation(dat, method = cor_fun)

# use kableExtra and psych to color significant cells
library(psych)
library(kableExtra)

dat <- mtcars[, c("vs", "hp", "gear")]

cor_fun <- function(dat) {
  # compute correlations and format them
  correlations <- data.frame(cor(dat))
  correlations <- datasummary_correlation_format(correlations, fmt = 2)

  # calculate pvalues using the `psych` package
  pvalues <- psych::corr.test(dat)$p

  # use `kableExtra::cell_spec` to color significant cells
  for (i in 1:nrow(correlations)) {
    for (j in 1:ncol(correlations)) {
      if (pvalues[i, j] < 0.05 && i != j) {
        correlations[i, j] <- cell_spec(correlations[i, j], background = "pink")
      }
    }
  }
  return(correlations)
}

# The `escape=FALSE` is important here!
datasummary_correlation(dat, method = cor_fun, escape = FALSE)

References

Arel-Bundock V (2022). “modelsummary: Data and Model Summaries in R.” Journal of Statistical Software, 103(1), 1-23. doi:10.18637/jss.v103.i01.'

Description

Mostly for internal use, but can be useful when users supply a function to the method argument of datasummary_correlation.

Usage

datasummary_correlation_format(
  x,
  fmt,
  leading_zero = FALSE,
  diagonal = NULL,
  upper_triangle = NULL,
  stars = FALSE
)

Arguments

Examples

library(modelsummary)

dat <- mtcars[, c("mpg", "hp", "disp")]

cor_fun <- function(x) {
  out <- cor(x, method = "kendall")
  datasummary_correlation_format(
    out,
    fmt = 2,
    upper_triangle = "x",
    diagonal = ".")
}

datasummary_correlation(dat, method = cor_fun)

Description

Convenience function to tabulate counts, cell percentages, and row/column percentages for categorical variables. See the Details section for a description of the internal design. For more complex cross tabulations, use datasummary directly. See the Details and Examples sections below, and the vignettes on the modelsummary website:

• https://modelsummary.com/

• https://modelsummary.com/vignettes/datasummary.html

Usage

datasummary_crosstab(
  formula,
  statistic = 1 ~ 1 + N + Percent("row"),
  data,
  output = getOption("modelsummary_output", default = "default"),
  fmt = 1,
  title = getOption("modelsummary_title", default = NULL),
  notes = getOption("modelsummary_notes", default = NULL),
  align = getOption("modelsummary_align", default = NULL),
  add_columns = getOption("modelsummary_add_columns", default = NULL),
  add_rows = getOption("modelsummary_add_rows", default = NULL),
  sparse_header = getOption("modelsummary_sparse_header", default = TRUE),
  escape = getOption("modelsummary_escape", default = TRUE),
  ...
)

Arguments

Details

datasummary_crosstab is a wrapper around the datasummary function. This wrapper works by creating a customized formula and by feeding it to datasummary. The customized formula comes in two parts.

First, we take a two-sided formula supplied by the formula argument. All variables of that formula are wrapped in a Factor() call to ensure that the variables are treated as categorical.

Second, the statistic argument gives a two-sided formula which specifies the statistics to include in the table. datasummary_crosstab modifies this formula automatically to include "clean" labels.

Finally, the formula and statistic formulas are combined into a single formula which is fed directly to the datasummary function to produce the table.

Variables in formula are automatically wrapped in Factor().

Version 2.0.0, kableExtra, and tinytable

Since version 2.0.0, modelsummary uses tinytable as its default table-drawing backend. Learn more at: https://vincentarelbundock.github.io/tinytable/",

Revert to kableExtra for one session:

options(modelsummary_factory_default = 'kableExtra') options(modelsummary_factory_latex = 'kableExtra') options(modelsummary_factory_html = 'kableExtra')

Global Options

The behavior of modelsummary can be modified by setting global options. In particular, most of the arguments for most of the package's functions cna be set using global options. For example:

• options(modelsummary_output = "modelsummary_list")

• options(modelsummary_statistic = '({conf.low}, {conf.high})')

• options(modelsummary_stars = TRUE)

Options not specific to given arguments are listed below.

Balance tables: summary functions

You can customize the summary statistics shown in datasummary_balance():

• options(modelsummary_balance_fn = list(Mean = Mean, "Std. Dev." = SD))

The list must be named, and each function should accept a numeric vector and return a single value. When this option is set and a weights column is present, weights are ignored with a warning.

Model labels: default column names

These global option changes the style of the default column headers:

• options(modelsummary_model_labels = "roman")

• options(modelsummary_model_labels_term = "Parameter")

• options(modelsummary_model_labels_group = "Group")

• options(modelsummary_model_labels_model = "Model")

The options above control the label shown in the stub columns for terms, groups, and model names. By default, those headers are blank. Setting any of these options to a character string inserts that label as the column header.

The supported styles are: "model", "arabic", "letters", "roman", "(arabic)", "(letters)", "(roman)"

Table-making packages

modelsummary supports 6 table-making packages: tinytable, kableExtra, gt, flextable, huxtable, and DT. Some of these packages have overlapping functionalities. To change the default backend used for a specific file format, you can use ' the options function:

options(modelsummary_factory_html = 'kableExtra') options(modelsummary_factory_word = 'huxtable') options(modelsummary_factory_png = 'gt') options(modelsummary_factory_latex = 'gt') options(modelsummary_factory_latex_tabular = 'kableExtra')

Table themes

Change the look of tables in an automated and replicable way, using the modelsummary theming functionality. See the vignette: https://modelsummary.com/vignettes/appearance.html

• modelsummary_theme_gt

• modelsummary_theme_kableExtra

• modelsummary_theme_huxtable

• modelsummary_theme_flextable

• modelsummary_theme_dataframe

Model extraction functions

modelsummary can use two sets of packages to extract information from statistical models: the easystats family (performance and parameters) and broom. By default, it uses easystats first and then falls back on broom in case of failure. You can change the order of priorities or include goodness-of-fit extracted by both packages by setting:

options(modelsummary_get = "easystats")

options(modelsummary_get = "broom")

options(modelsummary_get = "all")

The "all" option (default) means easystats then broom.

Formatting numeric entries

By default, LaTeX tables enclose all numeric entries in the ⁠\num{}⁠ command from the siunitx package. To prevent this behavior, or to enclose numbers in dollar signs (for LaTeX math mode), users can call:

options(modelsummary_format_numeric_latex = "plain")

options(modelsummary_format_numeric_latex = "mathmode")

A similar option can be used to display numerical entries using MathJax in HTML tables:

options(modelsummary_format_numeric_html = "mathjax")

LaTeX preamble

When creating LaTeX via the tinytable backend (default in version 2.0.0 and later), it is useful to include the following commands in the LaTeX preamble of your documents. These commands are automatically added to the preamble when compiling Rmarkdown or Quarto documents, except when the modelsummary() calls are cached.

\usepackage{tabularray}
\usepackage{float}
\usepackage{graphicx}
\usepackage[normalem]{ulem}
\UseTblrLibrary{booktabs}
\UseTblrLibrary{siunitx}
\newcommand{\tinytableTabularrayUnderline}[1]{\underline{#1}}
\newcommand{\tinytableTabularrayStrikeout}[1]{\sout{#1}}
\NewTableCommand{\tinytableDefineColor}[3]{\definecolor{#1}{#2}{#3}}

Examples

library(modelsummary)

# crosstab of two variables, showing counts, row percentages, and row/column totals
datasummary_crosstab(cyl ~ gear, data = mtcars)

# crosstab of two variables, showing counts only and no totals
datasummary_crosstab(cyl ~ gear, statistic = ~ N, data = mtcars)

# crosstab of three variables
  datasummary_crosstab(am * cyl ~ gear, data = mtcars)

# crosstab with two variables and column percentages
datasummary_crosstab(am ~ gear, statistic = ~ Percent("col"), data = mtcars)

References

Arel-Bundock V (2022). “modelsummary: Data and Model Summaries in R.” Journal of Statistical Software, 103(1), 1-23. doi:10.18637/jss.v103.i01.'

Description

Draw a table from a data.frame

Usage

datasummary_df(
  data,
  output = getOption("modelsummary_output", default = "default"),
  fmt = 2,
  align = getOption("modelsummary_align", default = NULL),
  hrule = getOption("modelsummary_hrule", default = NULL),
  title = getOption("modelsummary_title", default = NULL),
  notes = getOption("modelsummary_notes", default = NULL),
  add_rows = getOption("modelsummary_add_rows", default = NULL),
  add_columns = getOption("modelsummary_add_columns", default = NULL),
  escape = getOption("modelsummary_escape", default = TRUE),
  ...
)

Arguments

Version 2.0.0, kableExtra, and tinytable

Since version 2.0.0, modelsummary uses tinytable as its default table-drawing backend. Learn more at: https://vincentarelbundock.github.io/tinytable/",

Revert to kableExtra for one session:

options(modelsummary_factory_default = 'kableExtra') options(modelsummary_factory_latex = 'kableExtra') options(modelsummary_factory_html = 'kableExtra')

References

Arel-Bundock V (2022). “modelsummary: Data and Model Summaries in R.” Journal of Statistical Software, 103(1), 1-23. doi:10.18637/jss.v103.i01.'

Description

This function was inspired by the excellent skimr package for R. See the Details and Examples sections below, and the vignettes on the

Usage

datasummary_skim(
  data,
  output = getOption("modelsummary_output", default = "default"),
  type = getOption("modelsummary_type", default = "all"),
  fmt = 1,
  title = getOption("modelsummary_title", default = NULL),
  notes = getOption("modelsummary_notes", default = NULL),
  align = getOption("modelsummary_align", default = NULL),
  escape = getOption("modelsummary_escape", default = TRUE),
  by = getOption("modelsummary_by", default = NULL),
  fun_numeric = getOption("modelsummary_fun_numeric", default = list(Unique = NUnique,
    `Missing Pct.` = PercentMissing, Mean = Mean, SD = SD, Min = Min, Median = Median,
    Max = Max, Histogram = function(x) "")),
  ...
)

Arguments

Details

modelsummary website:

• https://modelsummary.com/

• https://modelsummary.com/vignettes/datasummary.html

Note that the default escape=TRUE may not be best for all formats, such as LaTeX/PDF.

Version 2.0.0, kableExtra, and tinytable

Since version 2.0.0, modelsummary uses tinytable as its default table-drawing backend. Learn more at: https://vincentarelbundock.github.io/tinytable/",

Revert to kableExtra for one session:

options(modelsummary_factory_default = 'kableExtra') options(modelsummary_factory_latex = 'kableExtra') options(modelsummary_factory_html = 'kableExtra')

Global Options

The behavior of modelsummary can be modified by setting global options. In particular, most of the arguments for most of the package's functions cna be set using global options. For example:

• options(modelsummary_output = "modelsummary_list")

• options(modelsummary_statistic = '({conf.low}, {conf.high})')

• options(modelsummary_stars = TRUE)

Options not specific to given arguments are listed below.

Balance tables: summary functions

You can customize the summary statistics shown in datasummary_balance():

• options(modelsummary_balance_fn = list(Mean = Mean, "Std. Dev." = SD))

The list must be named, and each function should accept a numeric vector and return a single value. When this option is set and a weights column is present, weights are ignored with a warning.

Model labels: default column names

These global option changes the style of the default column headers:

• options(modelsummary_model_labels = "roman")

• options(modelsummary_model_labels_term = "Parameter")

• options(modelsummary_model_labels_group = "Group")

• options(modelsummary_model_labels_model = "Model")

The options above control the label shown in the stub columns for terms, groups, and model names. By default, those headers are blank. Setting any of these options to a character string inserts that label as the column header.

The supported styles are: "model", "arabic", "letters", "roman", "(arabic)", "(letters)", "(roman)"

Table-making packages

modelsummary supports 6 table-making packages: tinytable, kableExtra, gt, flextable, huxtable, and DT. Some of these packages have overlapping functionalities. To change the default backend used for a specific file format, you can use ' the options function:

options(modelsummary_factory_html = 'kableExtra') options(modelsummary_factory_word = 'huxtable') options(modelsummary_factory_png = 'gt') options(modelsummary_factory_latex = 'gt') options(modelsummary_factory_latex_tabular = 'kableExtra')

Table themes

Change the look of tables in an automated and replicable way, using the modelsummary theming functionality. See the vignette: https://modelsummary.com/vignettes/appearance.html

• modelsummary_theme_gt

• modelsummary_theme_kableExtra

• modelsummary_theme_huxtable

• modelsummary_theme_flextable

• modelsummary_theme_dataframe

Model extraction functions

modelsummary can use two sets of packages to extract information from statistical models: the easystats family (performance and parameters) and broom. By default, it uses easystats first and then falls back on broom in case of failure. You can change the order of priorities or include goodness-of-fit extracted by both packages by setting:

options(modelsummary_get = "easystats")

options(modelsummary_get = "broom")

options(modelsummary_get = "all")

The "all" option (default) means easystats then broom.

Formatting numeric entries

By default, LaTeX tables enclose all numeric entries in the ⁠\num{}⁠ command from the siunitx package. To prevent this behavior, or to enclose numbers in dollar signs (for LaTeX math mode), users can call:

options(modelsummary_format_numeric_latex = "plain")

options(modelsummary_format_numeric_latex = "mathmode")

A similar option can be used to display numerical entries using MathJax in HTML tables:

options(modelsummary_format_numeric_html = "mathjax")

LaTeX preamble

When creating LaTeX via the tinytable backend (default in version 2.0.0 and later), it is useful to include the following commands in the LaTeX preamble of your documents. These commands are automatically added to the preamble when compiling Rmarkdown or Quarto documents, except when the modelsummary() calls are cached.

\usepackage{tabularray}
\usepackage{float}
\usepackage{graphicx}
\usepackage[normalem]{ulem}
\UseTblrLibrary{booktabs}
\UseTblrLibrary{siunitx}
\newcommand{\tinytableTabularrayUnderline}[1]{\underline{#1}}
\newcommand{\tinytableTabularrayStrikeout}[1]{\sout{#1}}
\NewTableCommand{\tinytableDefineColor}[3]{\definecolor{#1}{#2}{#3}}

References

Arel-Bundock V (2022). “modelsummary: Data and Model Summaries in R.” Journal of Statistical Software, 103(1), 1-23. doi:10.18637/jss.v103.i01.'

Examples

dat <- mtcars
dat$vs <- as.logical(dat$vs)
dat$cyl <- as.factor(dat$cyl)
datasummary_skim(dat)
datasummary_skim(dat, type = "categorical")

Description

A convenience function for use with a regression model or list of regression models. Returns a named list of models, where the names are the models' respective dependent variables. If the dependent variables are labelled, the labels are used as names. Pass your list of models to dvnames before sending to modelsummary to automatically get dependent variable-titled columns.

Usage

dvnames(models, number = FALSE, strip = FALSE, fill = "Model")

Arguments

Examples

m1 <- lm(mpg ~ hp, data = mtcars)
m2 <- lm(mpg ~ hp + wt, data = mtcars)

# Without dvnames, column names are (1) and (2)
modelsummary(list(m1, m2))

# With dvnames, they are "mpg" and "mpg"
modelsummary(dvnames(list(m1,m2)))

Description

Rounding with decimal digits in the fmt argument

Usage

fmt_decimal(digits = 3, pdigits = NULL, ...)

Arguments

Description

This function implements the suggestions of Astier & Wolak for the number of decimal digits to keep for coefficient estimates. The other statistics are rounded by fmt_significant().

Usage

fmt_equivalence(conf_level = 0.95, digits = 3, pdigits = NULL, ...)

Arguments

References

Astier, Nicolas, and Frank A. Wolak. Credible Numbers: A Procedure for Reporting Statistical Precision in Parameter Estimates. No. w32124. National Bureau of Economic Research, 2024.

Examples

library(modelsummary)
mod <- lm(mpg ~ hp, mtcars)

# Default equivalence-based formatting
modelsummary(mod, fmt = fmt_equivalence())

# alpha = 0.2
modelsummary(mod, fmt = fmt_equivalence(conf_level = .8))

# default equivalence, but with alternative significant digits for other statistics
modelsummary(mod, fmt = fmt_equivalence(digits = 5))

Description

Rounding using scientific notation

Usage

fmt_sci(digits = 3, ...)

Arguments

Description

The number of decimal digits to keep after the decimal is assessed

Usage

fmt_significant(digits = 3, ...)

Arguments

Description

Rounding with the sprintf() function in the fmt argument.

Usage

fmt_sprintf(fmt)

Arguments

Value

A function that takes an object and returns a string with the object formatted using sprintf.

Description

Rounding with decimal digits on a per-statistic basis in the fmt argument for modelsummary()

Usage

fmt_statistic(..., default = 3)

Arguments

Description

Rounding with decimal digits on a per-term basis in the fmt argument for modelsummary()

Usage

fmt_term(..., default = 3)

Arguments

Description

A unified approach to extract results from a wide variety of models.

Usage

get_estimates(
  model,
  conf_level = 0.95,
  vcov = NULL,
  shape = NULL,
  coef_rename = FALSE,
  ...
)

Arguments

Value

A dataframe with model estimates. The backend attribute of the returned object contains the backend that was used to extract parameters from the model. Moreover, for some models, get_estimates attaches useful attributes to the output. You can access this information by calling the attributes function: attributes(get_estimates(model)).

Description

A unified approach to extract results from a wide variety of models.

Usage

get_gof(model, gof_function = NULL, vcov_type = NULL, ...)

Arguments

Value

A dataframe with the goodness-o-fit statistics. The backend attribute indicates the backend used to extract them. Moreover, for some models get_gof attaches useful attributes to the output. You can access this information by calling the attributes function attributes(get_estimates(model)).

Description

By default, this data frame is passed to the 'gof_map' argument of the 'modelsummary' function. Users can modify this data frame to customize the list of statistics to display and their format. See example below.

Usage

gof_map

Format

data.frame with 4 columns of character data: raw, clean, fmt, omit

Examples

if (identical(Sys.getenv("pkgdown"),  "true")) {

library(modelsummary)
mod <- lm(wt ~ drat, data = mtcars)
gm <- modelsummary::gof_map
gm$omit[gm$raw == 'deviance'] <- FALSE
gm$fmt[gm$raw == 'r.squared'] <- "%.5f"
modelsummary(mod, gof_map = gm)
}

Description

Dot-Whisker plot of coefficient estimates with confidence intervals. For more information, see the Details and Examples sections below, and the vignettes on the modelsummary website: https://modelsummary.com/

• modelplot Vignette.

Usage

modelplot(
  models,
  conf_level = getOption("modelsummary_conf_level", default = 0.95),
  coef_map = getOption("modelsummary_coef_map", default = NULL),
  coef_omit = getOption("modelsummary_coef_omit", default = NULL),
  coef_rename = getOption("modelsummary_coef_rename", default = NULL),
  vcov = getOption("modelsummary_vcov", default = NULL),
  exponentiate = getOption("modelsummary_exponentiate", default = FALSE),
  add_rows = getOption("modelsummary_add_rows", default = NULL),
  facet = getOption("modelsummary_facet", default = FALSE),
  draw = getOption("modelsummary_draw", default = TRUE),
  background = getOption("modelsummary_background", default = NULL),
  ...
)

Arguments

Examples

library(modelsummary)

# single model
mod <- lm(hp ~ vs + drat, mtcars)
modelplot(mod)

# omit terms with string matches or regexes
modelplot(mod, coef_omit = 'Interc')

# rename, reorder and subset with 'coef_map'
cm <- c('vs' = 'V-shape engine',
  'drat' = 'Rear axle ratio')
modelplot(mod, coef_map = cm)

# several models
models <- list()
models[['Small model']] <- lm(hp ~ vs, mtcars)
models[['Medium model']] <- lm(hp ~ vs + factor(cyl), mtcars)
models[['Large model']] <- lm(hp ~ vs + drat + factor(cyl), mtcars)
modelplot(models)

# add_rows: add an empty reference category

mod <- lm(hp ~ factor(cyl), mtcars)

add_rows = data.frame(
  term = "factory(cyl)4",
  model = "(1)",
  estimate = NA)
attr(add_rows, "position") = 3
modelplot(mod, add_rows = add_rows)


# customize your plots with 'ggplot2' functions
library(ggplot2)

modelplot(models) +
  scale_color_brewer(type = 'qual') +
  theme_classic()

# pass arguments to 'geom_pointrange' through the ... ellipsis
modelplot(mod, color = 'red', size = 1, fatten = .5)

# add geoms to the background, behind geom_pointrange
b <- list(geom_vline(xintercept = 0, color = 'orange'),
  annotate("rect", alpha = .1,
    xmin = -.5, xmax = .5,
    ymin = -Inf, ymax = Inf),
  geom_point(aes(y = term, x = estimate), alpha = .3,
    size = 10, color = 'red', shape = 'square'))
modelplot(mod, background = b)

# logistic regression example
df <- as.data.frame(Titanic)
mod_titanic <- glm(
  Survived ~ Class + Sex,
  family = binomial,
  weight = Freq,
  data = df
)

# displaying odds ratio using a log scale
modelplot(mod_titanic, exponentiate = TRUE) +
  scale_x_log10() +
  xlab("Odds Ratios and 95% confidence intervals")

References

Arel-Bundock V (2022). “modelsummary: Data and Model Summaries in R.” Journal of Statistical Software, 103(1), 1-23. doi:10.18637/jss.v103.i01.'

Description

Create beautiful and customizable tables to summarize several statistical models side-by-side. This function supports dozens of statistical models, and it can produce tables in HTML, LaTeX, Word, Markdown, Typst, PDF, PowerPoint, Excel, RTF, JPG, or PNG. The appearance of the tables can be customized extensively by specifying the output argument, and by using functions from one of the supported table customization packages: tinytable, kableExtra, gt, flextable, huxtable, DT. For more information, see the Details and Examples sections below, and the vignettes on the modelsummary website: https://modelsummary.com/

• The modelsummary Vignette includes dozens of examples of tables with extensive customizations.

• The Appearance Vignette shows how to modify the look of tables.

Usage

modelsummary(
  models,
  output = getOption("modelsummary_output", default = "default"),
  fmt = getOption("modelsummary_fmt", default = 3),
  estimate = getOption("modelsummary_estimate", default = "estimate"),
  statistic = getOption("modelsummary_statistic", default = "std.error"),
  vcov = getOption("modelsummary_vcov", default = NULL),
  conf_level = getOption("modelsummary_conf_level", default = 0.95),
  exponentiate = getOption("modelsummary_exponentiate", default = FALSE),
  stars = getOption("modelsummary_stars", default = FALSE),
  shape = getOption("modelsummary_shape", default = term + statistic ~ model),
  coef_map = getOption("modelsummary_coef_map", default = NULL),
  coef_omit = getOption("modelsummary_coef_omit", default = NULL),
  coef_rename = getOption("modelsummary_coef_rename", default = FALSE),
  gof_map = getOption("modelsummary_gof_map", default = NULL),
  gof_omit = getOption("modelsummary_gof_omit", default = NULL),
  gof_function = getOption("modelsummary_gof_function", default = NULL),
  group_map = getOption("modelsummary_group_map", default = NULL),
  add_columns = getOption("modelsummary_add_columns", default = NULL),
  add_rows = getOption("modelsummary_add_rows", default = NULL),
  align = getOption("modelsummary_align", default = NULL),
  notes = getOption("modelsummary_notes", default = NULL),
  title = getOption("modelsummary_title", default = NULL),
  escape = getOption("modelsummary_escape", default = TRUE),
  ...
)

Arguments

Details

output

The modelsummary_list output is a lightweight format which can be used to save model results, so they can be fed back to modelsummary later to avoid extracting results again.

When a file name with a valid extension is supplied to the output argument, the table is written immediately to file. If you want to customize your table by post-processing it with an external package, you need to choose a different output format and saving mechanism. Unfortunately, the approach differs from package to package:

• tinytable: set output="tinytable", post-process your table, and use the tinytable::save_tt function.

• gt: set output="gt", post-process your table, and use the gt::gtsave function.

• kableExtra: set output to your destination format (e.g., "latex", "html", "markdown"), post-process your table, and use kableExtra::save_kable function.

vcov

To use a string such as "robust" or "HC0", your model must be supported by the sandwich package. This includes objects such as: lm, glm, survreg, coxph, mlogit, polr, hurdle, zeroinfl, and more.

NULL, "classical", "iid", and "constant" are aliases which do not modify uncertainty estimates and simply report the default standard errors stored in the model object.

One-sided formulas such as ~clusterid are passed to the sandwich::vcovCL function.

Matrices and functions producing variance-covariance matrices are first passed to lmtest. If this does not work, modelsummary attempts to take the square root of the diagonal to adjust "std.error", but the other uncertainty estimates are not be adjusted.

Numeric vectors are formatted according to fmt and placed in brackets. Character vectors printed as given, without parentheses.

If your model type is supported by the lmtest package, the vcov argument will try to use that package to adjust all the uncertainty estimates, including "std.error", "statistic", "p.value", and "conf.int". If your model is not supported by lmtest, only the "std.error" will be adjusted by, for example, taking the square root of the matrix's diagonal.

Value

a regression table in a format determined by the output argument. The backend attribute includes the backend used to extract estimates and goodness-of-fit measure.

Version 2.0.0, kableExtra, and tinytable

Since version 2.0.0, modelsummary uses tinytable as its default table-drawing backend. Learn more at: https://vincentarelbundock.github.io/tinytable/",

Revert to kableExtra for one session:

options(modelsummary_factory_default = 'kableExtra') options(modelsummary_factory_latex = 'kableExtra') options(modelsummary_factory_html = 'kableExtra')

Global Options

The behavior of modelsummary can be modified by setting global options. In particular, most of the arguments for most of the package's functions cna be set using global options. For example:

• options(modelsummary_output = "modelsummary_list")

• options(modelsummary_statistic = '({conf.low}, {conf.high})')

• options(modelsummary_stars = TRUE)

Options not specific to given arguments are listed below.

Balance tables: summary functions

You can customize the summary statistics shown in datasummary_balance():

• options(modelsummary_balance_fn = list(Mean = Mean, "Std. Dev." = SD))

The list must be named, and each function should accept a numeric vector and return a single value. When this option is set and a weights column is present, weights are ignored with a warning.

Model labels: default column names

These global option changes the style of the default column headers:

• options(modelsummary_model_labels = "roman")

• options(modelsummary_model_labels_term = "Parameter")

• options(modelsummary_model_labels_group = "Group")

• options(modelsummary_model_labels_model = "Model")

The options above control the label shown in the stub columns for terms, groups, and model names. By default, those headers are blank. Setting any of these options to a character string inserts that label as the column header.

The supported styles are: "model", "arabic", "letters", "roman", "(arabic)", "(letters)", "(roman)"

Table-making packages

modelsummary supports 6 table-making packages: tinytable, kableExtra, gt, flextable, huxtable, and DT. Some of these packages have overlapping functionalities. To change the default backend used for a specific file format, you can use ' the options function:

options(modelsummary_factory_html = 'kableExtra') options(modelsummary_factory_word = 'huxtable') options(modelsummary_factory_png = 'gt') options(modelsummary_factory_latex = 'gt') options(modelsummary_factory_latex_tabular = 'kableExtra')

Table themes

Change the look of tables in an automated and replicable way, using the modelsummary theming functionality. See the vignette: https://modelsummary.com/vignettes/appearance.html

• modelsummary_theme_gt

• modelsummary_theme_kableExtra

• modelsummary_theme_huxtable

• modelsummary_theme_flextable

• modelsummary_theme_dataframe

Model extraction functions

modelsummary can use two sets of packages to extract information from statistical models: the easystats family (performance and parameters) and broom. By default, it uses easystats first and then falls back on broom in case of failure. You can change the order of priorities or include goodness-of-fit extracted by both packages by setting:

options(modelsummary_get = "easystats")

options(modelsummary_get = "broom")

options(modelsummary_get = "all")

The "all" option (default) means easystats then broom.

Formatting numeric entries

By default, LaTeX tables enclose all numeric entries in the ⁠\num{}⁠ command from the siunitx package. To prevent this behavior, or to enclose numbers in dollar signs (for LaTeX math mode), users can call:

options(modelsummary_format_numeric_latex = "plain")

options(modelsummary_format_numeric_latex = "mathmode")

A similar option can be used to display numerical entries using MathJax in HTML tables:

options(modelsummary_format_numeric_html = "mathjax")

LaTeX preamble

When creating LaTeX via the tinytable backend (default in version 2.0.0 and later), it is useful to include the following commands in the LaTeX preamble of your documents. These commands are automatically added to the preamble when compiling Rmarkdown or Quarto documents, except when the modelsummary() calls are cached.

\usepackage{tabularray}
\usepackage{float}
\usepackage{graphicx}
\usepackage[normalem]{ulem}
\UseTblrLibrary{booktabs}
\UseTblrLibrary{siunitx}
\newcommand{\tinytableTabularrayUnderline}[1]{\underline{#1}}
\newcommand{\tinytableTabularrayStrikeout}[1]{\sout{#1}}
\NewTableCommand{\tinytableDefineColor}[3]{\definecolor{#1}{#2}{#3}}

Parallel computation

It can take a long time to compute and extract summary statistics from certain models (e.g., Bayesian). In those cases, users can parallelize the process. Since parallelization occurs at the model level, no speedup is available for tables with a single model. Users on mac or linux can launch parallel computation using the built-in parallel package. All they need to do is supply a mc.cores argument which will be pushed forward to the parallel::mclapply function:

modelsummary(model_list, mc.cores = 5)

All users can also use the future.apply package to parallelize model summaries. For example, to use 4 cores to extract results:

library(future.apply)
plan(multicore, workers = 4)
options("modelsummary_future" = TRUE)
modelsummary(model_list)

Note that the "multicore" plan only parallelizes under mac or linux. Windows users can use plan(multisession) instead. However, note that the first time modelsummary() is called under multisession can be a fair bit longer, because of extra costs in passing data to and loading required packages on to workers. Subsequent calls to modelsummary() will often be much faster.

Some users have reported difficult to reproduce errors when using the future package with some packages. The future parallelization in modelsummary can be disabled by calling:

options("modelsummary_future" = FALSE)

References

Arel-Bundock V (2022). “modelsummary: Data and Model Summaries in R.” Journal of Statistical Software, 103(1), 1-23. doi:10.18637/jss.v103.i01.'

Examples

# The `modelsummary` website includes \emph{many} examples and tutorials:
# https://modelsummary.com

library(modelsummary)

# load data and estimate models
utils::data(trees)
models <- list()
models[["Bivariate"]] <- lm(Girth ~ Height, data = trees)
models[["Multivariate"]] <- lm(Girth ~ Height + Volume, data = trees)

# simple table
modelsummary(models)

# statistic
modelsummary(models, statistic = NULL)

modelsummary(models, statistic = "p.value")

modelsummary(models, statistic = "statistic")

modelsummary(models, statistic = "conf.int", conf_level = 0.99)

modelsummary(models, statistic = c(
  "t = {statistic}",
  "se = {std.error}",
  "conf.int"))

# estimate
modelsummary(models,
  statistic = NULL,
  estimate = "{estimate} [{conf.low}, {conf.high}]")

modelsummary(models,
  estimate = c(
    "{estimate}{stars}",
    "{estimate} ({std.error})"))

# vcov
modelsummary(models, vcov = "robust")

modelsummary(models, vcov = list("classical", "stata"))

modelsummary(models, vcov = sandwich::vcovHC)

modelsummary(models,
  vcov = list(stats::vcov, sandwich::vcovHC))

modelsummary(models,
  vcov = list(
    c("(Intercept)" = "", "Height" = "!"),
    c("(Intercept)" = "", "Height" = "!", "Volume" = "!!")))

# vcov with custom names
modelsummary(
  models,
  vcov = list(
    "Stata Corp" = "stata",
    "Newey Lewis & the News" = "NeweyWest"))

# fmt
mod <- lm(mpg ~ hp + drat + qsec, data = mtcars)

modelsummary(mod, fmt = 3)

modelsummary(mod, fmt = fmt_significant(3))

modelsummary(mod, fmt = NULL)

modelsummary(mod, fmt = fmt_decimal(4))

modelsummary(mod, fmt = fmt_sprintf("%.5f"))

modelsummary(mod, fmt = fmt_statistic(estimate = 4, conf.int = 1), statistic = "conf.int")

modelsummary(mod, fmt = fmt_term(hp = 4, drat = 1, default = 2))

m <- lm(mpg ~ I(hp * 1000) + drat, data = mtcars)
f <- function(x) format(x, digits = 3, nsmall = 2, scientific = FALSE, trim = TRUE)
modelsummary(m, fmt = f, gof_map = NA)

# coef_rename
modelsummary(models, coef_rename = c("Volume" = "Large", "Height" = "Tall"))

modelsummary(models, coef_rename = toupper)

modelsummary(models, coef_rename = coef_rename)

# coef_rename = TRUE for variable labels
datlab <- mtcars
datlab$cyl <- factor(datlab$cyl)
attr(datlab$hp, "label") <- "Horsepower"
attr(datlab$cyl, "label") <- "Cylinders"
modlab <- lm(mpg ~ hp * drat + cyl, data = datlab)
modelsummary(modlab, coef_rename = TRUE)

# coef_rename: unnamed vector of length equal to the number of terms in the final table
m <- lm(hp ~ mpg + factor(cyl), data = mtcars)
modelsummary(m, coef_omit = -(3:4), coef_rename = c("Cyl 6", "Cyl 8"))

# coef_map
modelsummary(models, coef_map = c("Volume" = "Large", "Height" = "Tall"))

modelsummary(models, coef_map = c("Volume", "Height"))

# coef_omit: omit the first and second coefficients
modelsummary(models, coef_omit = 1:2)

# coef_omit: omit coefficients matching one substring
modelsummary(models, coef_omit = "ei", gof_omit = ".*")

# coef_omit: omit a specific coefficient
modelsummary(models, coef_omit = "^Volume$", gof_omit = ".*")

# coef_omit: omit coefficients matching either one of two substring
# modelsummary(models, coef_omit = "ei|rc", gof_omit = ".*")

# coef_omit: keep coefficients starting with a substring (using a negative lookahead)
# modelsummary(models, coef_omit = "^(?!Vol)", gof_omit = ".*")

# coef_omit: keep coefficients matching a substring
modelsummary(models, coef_omit = "^(?!.*ei|.*pt)", gof_omit = ".*")

# shape: multinomial model
library(nnet)
multi <- multinom(factor(cyl) ~ mpg + hp, data = mtcars, trace = FALSE)

# shape: term names and group ids in rows, models in columns
modelsummary(multi, shape = response ~ model)

# shape: term names and group ids in rows in a single column
modelsummary(multi, shape = term:response ~ model)

# shape: term names in rows and group ids in columns
modelsummary(multi, shape = term ~ response:model)

# shape = "rcollapse"
panels <- list(
  "Panel A: MPG" = list(
    "A" = lm(mpg ~ hp, data = mtcars),
    "B" = lm(mpg ~ hp + factor(gear), data = mtcars)),
  "Panel B: Displacement" = list(
    "A" = lm(disp ~ hp, data = mtcars),
    "C" = lm(disp ~ hp + factor(gear), data = mtcars))
)

# shape = "cbind"
modelsummary(panels, shape = "cbind")

modelsummary(
  panels,
  shape = "rbind",
  gof_map = c("nobs", "r.squared"))

# title
modelsummary(models, title = "This is the title")

# title with LaTeX label (for numbering and referencing)
modelsummary(models, title = "This is the title \\label{tab:description}", escape = FALSE)

# add_rows
rows <- tibble::tribble(
  ~term, ~Bivariate, ~Multivariate,
  "Empty row", "-", "-",
  "Another empty row", "?", "?")
attr(rows, "position") <- c(1, 3)
modelsummary(models, add_rows = rows)

attr(rows, "position") <- "gof_start"
modelsummary(models, add_rows = rows)

# notes
modelsummary(models, notes = list("A first note", "A second note"))

# gof_map: tribble
library(tibble)
gm <- tribble(
  ~raw, ~clean, ~fmt,
  "r.squared", "R Squared", 5)
modelsummary(models, gof_map = gm)

# gof_map: list of lists
f <- function(x) format(round(x, 3), big.mark = ",")
gm <- list(
  list("raw" = "nobs", "clean" = "N", "fmt" = f),
  list("raw" = "AIC", "clean" = "aic", "fmt" = f))
modelsummary(models, gof_map = gm)

Description

Update modelsummary and its dependencies to the latest R-Universe or CRAN versions. The R session needs to be restarted after install.

Usage

update_modelsummary(source = "development")

Arguments