| Title: | An Arsenal of 'R' Functions for Large-Scale Statistical Summaries |
|---|---|
| Description: | An Arsenal of 'R' functions for large-scale statistical summaries, which are streamlined to work within the latest reporting tools in 'R' and 'RStudio' and which use formulas and versatile summary statistics for summary tables and models. The primary functions include tableby(), a Table-1-like summary of multiple variable types 'by' the levels of one or more categorical variables; paired(), a Table-1-like summary of multiple variable types paired across two time points; modelsum(), which performs simple model fits on one or more endpoints for many variables (univariate or adjusted for covariates); freqlist(), a powerful frequency table across many categorical variables; comparedf(), a function for comparing data.frames; and write2(), a function to output tables to a document. |
| Authors: | Ethan Heinzen [aut, cre], Jason Sinnwell [aut], Elizabeth Atkinson [aut], Tina Gunderson [aut], Gregory Dougherty [aut], Patrick Votruba [ctb], Ryan Lennon [ctb], Andrew Hanson [ctb], Krista Goergen [ctb], Emily Lundt [ctb], Brendan Broderick [ctb], Maddie McCullough [art] |
| Maintainer: | Ethan Heinzen <[email protected]> |
| License: | GPL (>= 2) |
| Version: | 3.6.4.0000 |
| Built: | 2025-02-23 06:18:36 UTC |
| Source: | https://github.com/mayoverse/arsenal |
The not-in operator for R.
x %nin% tablex %nin% table
x |
vector or |
table |
vector or |
The negation of %in%.
Raymond Moore
1 %nin% 2:10 c("a", "b") %nin% c("a", "c", "d")1 %nin% 2:10 c("a", "b") %nin% c("a", "c", "d")
An Arsenal of 'R' functions for large-scale statistical summaries, which are streamlined to work within the latest reporting tools in 'R' and 'RStudio' and which use formulas and versatile summary statistics for summary tables and models.
The package download, NEWS, and README are available on CRAN: https://cran.r-project.org/package=arsenal
Below are listed some of the most widely used functions available in arsenal:
tableby: Summary statistics of a set of independent variables by a categorical variable.
paired: Summary statistics of a set of independent variables paired across two timepoints.
modelsum: Fit models over each of a set of independent variables with a response variable.
freqlist: Approximate the output from SAS's PROC FREQ procedure when using
the /list option of the TABLE statement.
comparedf: Compare two data.frames and report any differences between them,
much like SAS's PROC COMPARE procedure.
write2word, write2html, write2pdf: Functions to output
tables to a single Word, HTML, or PDF document.
write2: Functions to output tables to a single document.
(Also the S3 backbone behind the write2* functions.)
keep.labels: Keep the 'label' attribute on an R object when subsetting.
formulize: A shortcut to generate one-, two-, or many-sided formulas.
mdy.Date and Date.mdy: Convert numeric dates for month, day, and year to Date object, and vice versa.
is.Date: Test if an object is a date.
%nin%: Test for "not in".
allNA and includeNA: some useful functions for dealing with NAs.
mockstudy: Mock study data for examples.
library(arsenal)library(arsenal)
arsenal tables with common structurearsenal tables with common structure
has_strata(x) ## S3 method for class 'arsenal_table' x[i, j, ...] ## S3 method for class 'tableby' x[i, j, ...] ## S3 method for class 'arsenal_table' labels(object, ...) ## S3 replacement method for class 'arsenal_table' labels(x) <- value ## S3 method for class 'arsenal_table' print(x, ...) ## S3 method for class 'arsenal_table' merge(x, y, all = FALSE, all.x = all, all.y = all, ...) ## S3 method for class 'freqlist' merge(x, y, all = TRUE, ...) ## S3 method for class 'summary.arsenal_table' print( x, ..., format = if (!is.null(x$text) && x$text %in% c("html", "latex")) x$text else "markdown", escape = x$text %nin% c("html", "latex"), width = NULL, min.split = NULL )has_strata(x) ## S3 method for class 'arsenal_table' x[i, j, ...] ## S3 method for class 'tableby' x[i, j, ...] ## S3 method for class 'arsenal_table' labels(object, ...) ## S3 replacement method for class 'arsenal_table' labels(x) <- value ## S3 method for class 'arsenal_table' print(x, ...) ## S3 method for class 'arsenal_table' merge(x, y, all = FALSE, all.x = all, all.y = all, ...) ## S3 method for class 'freqlist' merge(x, y, all = TRUE, ...) ## S3 method for class 'summary.arsenal_table' print( x, ..., format = if (!is.null(x$text) && x$text %in% c("html", "latex")) x$text else "markdown", escape = x$text %nin% c("html", "latex"), width = NULL, min.split = NULL )
x, y, object
|
An object of class |
i, j
|
A vector to index |
... |
Other arguments (only used in |
value |
A list of new labels. |
all, all.x, all.y
|
Logicals, denoting which terms to keep if not all are in common. |
format |
Passed to |
escape |
Passed to |
width, min.split
|
Passed to |
arsenal
Details about defunct functions in arsenal
x, y
|
See |
... |
Other arguments. |
comparison.control was renamed to comparedf.control in version 3.0.0.
compare.data.frame was renamed to comparedf in version 3.0.0.
length.tableby was removed in version 2.0.0.
includeNA.character and includeNA.numeric were removed in version 2.0.0 and
replaced with a default method.
rangeTime was removed in version 1.5.0.
Convert freqlist object to a data.frame.
## S3 method for class 'freqlist' as.data.frame(x, ..., labelTranslations = NULL, list.ok = FALSE)## S3 method for class 'freqlist' as.data.frame(x, ..., labelTranslations = NULL, list.ok = FALSE)
x |
An object of class |
... |
Arguments to pass to |
labelTranslations |
A named list (or vector) where the name is the label in the
output to be replaced in the pretty rendering by the character string
value for the named element of the list, e.g., |
list.ok |
If the object has multiple by-variables, is it okay to return a list of data.frames instead of a single data.frame?
If |
A data.frame corresponding to the freqlist object.
Coerce a modelsum object to a data.frame.
## S3 method for class 'modelsum' as.data.frame(x, ..., labelTranslations = NULL, list.ok = FALSE)## S3 method for class 'modelsum' as.data.frame(x, ..., labelTranslations = NULL, list.ok = FALSE)
x |
A |
... |
Arguments to pass to |
labelTranslations |
A named list (or vector) where the name is the label in the
output to be replaced in the pretty rendering by the character string
value for the named element of the list, e.g., |
list.ok |
If the object has multiple by-variables, is it okay to return a list of data.frames instead of a single data.frame?
If |
A data.frame.
Ethan Heinzen, based on code originally by Greg Dougherty
Coerce a tableby object to a data.frame.
## S3 method for class 'tableby' as.data.frame(x, ..., labelTranslations = NULL, list.ok = FALSE)## S3 method for class 'tableby' as.data.frame(x, ..., labelTranslations = NULL, list.ok = FALSE)
x |
A |
... |
Arguments to pass to |
labelTranslations |
A named list (or vector) where the name is the label in the
output to be replaced in the pretty rendering by the character string
value for the named element of the list, e.g., |
list.ok |
If the object has multiple by-variables, is it okay to return a list of data.frames instead of a single data.frame?
If |
A data.frame.
Ethan Heinzen, based on code originally by Greg Dougherty
Compare two data.frames and report any differences between them,
much like SAS's PROC COMPARE procedure.
comparedf(x, y, by = NULL, by.x = by, by.y = by, control = NULL, ...) ## S3 method for class 'comparedf' print(x, ...)comparedf(x, y, by = NULL, by.x = by, by.y = by, control = NULL, ...) ## S3 method for class 'comparedf' print(x, ...)
x, y
|
A data.frame to compare |
by, by.x, by.y
|
Which variables are IDs to merge the two data.frames?
If set to |
control |
A list of control parameters from |
... |
Other arguments, passed to |
Ethan Heinzen, adapted from code from Andrew Hanson
summary.comparedf, comparedf.control,
diffs, n.diffs, n.diff.obs
df1 <- data.frame(id = paste0("person", 1:3), a = c("a", "b", "c"), b = c(1, 3, 4), c = c("f", "e", "d"), row.names = paste0("rn", 1:3), stringsAsFactors = FALSE) df2 <- data.frame(id = paste0("person", 3:1), a = c("c", "b", "a"), b = c(1, 3, 4), d = paste0("rn", 1:3), row.names = paste0("rn", c(1,3,2)), stringsAsFactors = FALSE) summary(comparedf(df1, df2)) summary(comparedf(df1, df2, by = "id")) summary(comparedf(df1, df2, by = "row.names"))df1 <- data.frame(id = paste0("person", 1:3), a = c("a", "b", "c"), b = c(1, 3, 4), c = c("f", "e", "d"), row.names = paste0("rn", 1:3), stringsAsFactors = FALSE) df2 <- data.frame(id = paste0("person", 3:1), a = c("c", "b", "a"), b = c(1, 3, 4), d = paste0("rn", 1:3), row.names = paste0("rn", c(1,3,2)), stringsAsFactors = FALSE) summary(comparedf(df1, df2)) summary(comparedf(df1, df2, by = "id")) summary(comparedf(df1, df2, by = "row.names"))
comparedf functionControl tolerance definitions for the comparedf function.
comparedf.control( tol.logical = "none", tol.num = c("absolute", "percent", "pct"), tol.num.val = sqrt(.Machine$double.eps), int.as.num = FALSE, tol.char = c("none", "trim", "case", "both"), tol.factor = c("none", "levels", "labels"), factor.as.char = FALSE, tol.date = "absolute", tol.date.val = 0, tol.other = "none", tol.vars = "none", max.print.vars = NA, max.print.obs = NA, max.print.diffs.per.var = 10, max.print.diffs = 50, max.print.attrs = NA, ..., max.print.diff = 10 )comparedf.control( tol.logical = "none", tol.num = c("absolute", "percent", "pct"), tol.num.val = sqrt(.Machine$double.eps), int.as.num = FALSE, tol.char = c("none", "trim", "case", "both"), tol.factor = c("none", "levels", "labels"), factor.as.char = FALSE, tol.date = "absolute", tol.date.val = 0, tol.other = "none", tol.vars = "none", max.print.vars = NA, max.print.obs = NA, max.print.diffs.per.var = 10, max.print.diffs = 50, max.print.attrs = NA, ..., max.print.diff = 10 )
tol.logical, tol.num, tol.char, tol.factor, tol.date, tol.other
|
A function or one of the shortcut character strings or a list thereof, denoting the tolerance function to use for a given data type. See "details", below. |
tol.num.val |
Numeric; maximum value of differences allowed in numerics (fed to the function given in |
int.as.num |
Logical; should integers be coerced to numeric before comparison? Default FALSE. |
factor.as.char |
Logical; should factors be coerced to character before comparison? Default FALSE. |
tol.date.val |
Numeric; maximum value of differences allowed in dates (fed to the function given in |
tol.vars |
Either |
max.print.vars |
Integer denoting maximum number of variables to report in the "variables not shared" and "variables not compared"
output. |
max.print.obs |
Integer denoting maximum number of not-shared observations to report. |
max.print.diffs.per.var, max.print.diffs
|
Integers denoting the maximum number of differences to report for each variable or overall.
|
max.print.attrs |
Integers denoting the maximum number of non-identical attributes to report. |
... |
Other arguments (not in use at this time). |
max.print.diff |
Deprecated. |
The following character strings are accepted:
tol.logical = "none": compare logicals exactly as they are.
tol.num = "absolute": compare absolute differences in numerics.
tol.num = "percent", tol.num = "pct" compare percent differences in numerics.
tol.char = "none": compare character strings exactly as they are.
tol.char = "trim": left-justify and trim all trailing white space.
tol.char = "case": allow differences in upper/lower case.
tol.char = "both": combine "trim" and "case".
tol.factor = "none": match both character labels and numeric levels.
tol.factor = "levels": match only the numeric levels.
tol.factor = "labels": match only the labels.
tol.date = "absolute": compare absolute differences in dates.
tol.other = "none": expect objects of other classes to be exactly identical.
A list with names mapped to x can be used to specify tolerances by variable. One unnamed element is supported
as the default.
tol.vars: If not set to "none" (the default) or a named vector,
the tol.vars argument is a character vector denoting equivalence classes
for the characters in the variable names. A single character in this vector means to replace that character
with "". All other strings in this vector are split by character and replaced by the first character in the string.
E.g., a character vector c("._", "aA", " ") would denote that the dot and underscore are equivalent (to be translated to a dot),
that "a" and "A" are equivalent (to be translated to "a"), and that spaces should be removed.
The special character string "case" in this vector is the same as specifying paste0(letters, LETTERS).
A list containing the necessary parameters for the comparedf function.
Ethan Heinzen
comparedf, comparedf.tolerances, summary.comparedf
cntl <- comparedf.control( tol.num = "pct", # calculate percent differences tol.vars = c("case", # ignore case "._", # set all underscores to dots. "e") # remove all letter e's ) cntl <- comparedf.control(tol.char = list( "none", # the default x1 = "case", # be case-insensitive for the variable "x1" x2 = function(x, y) tol.NA(x, y, x != y | y == "NA") # a custom-defined tolerance ))cntl <- comparedf.control( tol.num = "pct", # calculate percent differences tol.vars = c("case", # ignore case "._", # set all underscores to dots. "e") # remove all letter e's ) cntl <- comparedf.control(tol.char = list( "none", # the default x1 = "case", # be case-insensitive for the variable "x1" x2 = function(x, y) tol.NA(x, y, x != y | y == "NA") # a custom-defined tolerance ))
comparedf tolerancesInternal functions defining tolerances for the comparedf.control function.
To create your own tolerance definitions, see the vignette.
tol.NA(x, y, idx) tol.num.absolute(x, y, tol) tol.num.percent(x, y, tol) tol.num.pct(x, y, tol) tol.factor.none(x, y) tol.factor.levels(x, y) tol.factor.labels(x, y) tol.char.both(x, y) tol.char.case(x, y) tol.char.trim(x, y) tol.char.none(x, y) tol.date.absolute(x, y, tol) tol.logical.none(x, y) tol.other.none(x, y)tol.NA(x, y, idx) tol.num.absolute(x, y, tol) tol.num.percent(x, y, tol) tol.num.pct(x, y, tol) tol.factor.none(x, y) tol.factor.levels(x, y) tol.factor.labels(x, y) tol.char.both(x, y) tol.char.case(x, y) tol.char.trim(x, y) tol.char.none(x, y) tol.date.absolute(x, y, tol) tol.logical.none(x, y) tol.other.none(x, y)
x, y
|
vectors of the appropriate lengths and types. |
idx |
A logical vector of appropriate length. |
tol |
A numeric tolerance |
tol.NA takes as differences between two vectors any elements which are NA in one but not the other,
or which are non-NA in both and TRUE in idx. It is useful for handling NAs in custom tolerance functions.
A logical vector of length equal to that of x and y, where TRUE denotes a
difference between x and y, and FALSE denotes no difference between x and y.
Ethan Heinzen
Extract differences (diffs()), number of differences (n.diffs()),
or number of not-shared observations (n.diff.obs()) from a comparedf object.
n.diff.obs(object, ...) ## S3 method for class 'comparedf' n.diff.obs(object, ...) ## S3 method for class 'summary.comparedf' n.diff.obs(object, ...) n.diffs(object, ...) ## S3 method for class 'comparedf' n.diffs(object, ...) ## S3 method for class 'summary.comparedf' n.diffs(object, ...) diffs(object, ...) ## S3 method for class 'comparedf' diffs( object, what = c("differences", "observations"), vars = NULL, ..., by.var = FALSE ) ## S3 method for class 'summary.comparedf' diffs( object, what = c("differences", "observations"), vars = NULL, ..., by.var = FALSE )n.diff.obs(object, ...) ## S3 method for class 'comparedf' n.diff.obs(object, ...) ## S3 method for class 'summary.comparedf' n.diff.obs(object, ...) n.diffs(object, ...) ## S3 method for class 'comparedf' n.diffs(object, ...) ## S3 method for class 'summary.comparedf' n.diffs(object, ...) diffs(object, ...) ## S3 method for class 'comparedf' diffs( object, what = c("differences", "observations"), vars = NULL, ..., by.var = FALSE ) ## S3 method for class 'summary.comparedf' diffs( object, what = c("differences", "observations"), vars = NULL, ..., by.var = FALSE )
object |
An object of class |
... |
Other arguments (not in use at this time). |
what |
Should differences or the not-shared observations be returned? |
vars |
A character vector of variable names to subset the results to. |
by.var |
Logical: should the number of differences by variable be reported, or should all differences be reported (the default). |
Ethan Heinzen
A shortcut to generate one-, two-, or many-sided formulas from vectors of variable names.
formulize( y = "", x, ..., data = NULL, collapse = "+", collapse.y = collapse, escape = FALSE )formulize( y = "", x, ..., data = NULL, collapse = "+", collapse.y = collapse, escape = FALSE )
y, x, ...
|
Character vectors, names, or calls to be collapsed (by |
data |
An R object with non-null column names. |
collapse |
How should terms be collapsed? Default is addition. |
collapse.y |
How should the y-terms be collapsed? Default is addition. Also accepts the special string "list", which combines them into a multiple-left-hand-side formula, for use in other functions. |
escape |
A logical indicating whether character vectors should be coerced to names (that is, whether names with spaces should be surrounded with backticks or not) |
Ethan Heinzen
## two-sided formula f1 <- formulize("y", c("x1", "x2", "x3")) ## one-sided formula f2 <- formulize(x = c("x1", "x2", "x3")) ## multi-sided formula f3 <- formulize("y", c("x1", "x2", "x3"), c("z1", "z2"), "w1") ## can use numerics for column names data(mockstudy) f4 <- formulize(y = 1, x = 2:4, data = mockstudy) ## mix and match f5 <- formulize(1, c("x1", "x2", "x3"), data = mockstudy) ## get an interaction f6 <- formulize("y", c("x1*x2", "x3")) ## get only interactions f7 <- formulize("y", c("x1", "x2", "x3"), collapse = "*") ## no intercept f8 <- formulize("y", "x1 - 1") f9 <- formulize("y", c("x1", "x2", "-1")) ## LHS as a list to use in arsenal functions f10 <- formulize(c("y1", "y2", "y3"), c("x", "z"), collapse.y = "list") ## use in an lm f11 <- formulize(2, 3:4, data = mockstudy) summary(lm(f11, data = mockstudy)) ## using non-syntactic names or calls (like reformulate example) f12 <- formulize(as.name("+-"), c("`P/E`", "`% Growth`")) f12 <- formulize("+-", c("P/E", "% Growth"), escape = TRUE) f <- Surv(ft, case) ~ a + b f13 <- formulize(f[[2]], f[[3]])## two-sided formula f1 <- formulize("y", c("x1", "x2", "x3")) ## one-sided formula f2 <- formulize(x = c("x1", "x2", "x3")) ## multi-sided formula f3 <- formulize("y", c("x1", "x2", "x3"), c("z1", "z2"), "w1") ## can use numerics for column names data(mockstudy) f4 <- formulize(y = 1, x = 2:4, data = mockstudy) ## mix and match f5 <- formulize(1, c("x1", "x2", "x3"), data = mockstudy) ## get an interaction f6 <- formulize("y", c("x1*x2", "x3")) ## get only interactions f7 <- formulize("y", c("x1", "x2", "x3"), collapse = "*") ## no intercept f8 <- formulize("y", "x1 - 1") f9 <- formulize("y", c("x1", "x2", "-1")) ## LHS as a list to use in arsenal functions f10 <- formulize(c("y1", "y2", "y3"), c("x", "z"), collapse.y = "list") ## use in an lm f11 <- formulize(2, 3:4, data = mockstudy) summary(lm(f11, data = mockstudy)) ## using non-syntactic names or calls (like reformulate example) f12 <- formulize(as.name("+-"), c("`P/E`", "`% Growth`")) f12 <- formulize("+-", c("P/E", "% Growth"), escape = TRUE) f <- Surv(ft, case) ~ a + b f13 <- formulize(f[[2]], f[[3]])
freqlist functionControl test and summary settings for the freqlist function.
freq.control( sparse = FALSE, single = FALSE, dupLabels = FALSE, digits.count = 0L, digits.pct = 2L, ..., digits = NULL )freq.control( sparse = FALSE, single = FALSE, dupLabels = FALSE, digits.count = 0L, digits.pct = 2L, ..., digits = NULL )
sparse |
a logical value indicating whether to keep rows with counts of zero.
The default is |
single |
logical, indicating whether to collapse results created using a strata variable into a single table for printing |
dupLabels |
logical: should labels which are the same as the row above be printed? The default ( |
digits.count |
Number of decimal places for count values. |
digits.pct |
Number of decimal places for percents. |
... |
additional arguments. |
digits |
A deprecated argument |
A list with settings to be used within the freqlist function.
Ethan Heinzen
freqlist, summary.freqlist, freqlist.internal
Approximate the output from SAS's PROC FREQ procedure when using the /list option of the TABLE statement.
freqlist(object, ...) ## S3 method for class 'table' freqlist( object, na.options = c("include", "showexclude", "remove"), strata = NULL, labelTranslations = NULL, control = NULL, ... ) ## S3 method for class 'formula' freqlist( formula, data, subset, na.action, na.options = c("include", "showexclude", "remove"), strata = NULL, labelTranslations = NULL, control = NULL, addNA, exclude, drop.unused.levels, ... )freqlist(object, ...) ## S3 method for class 'table' freqlist( object, na.options = c("include", "showexclude", "remove"), strata = NULL, labelTranslations = NULL, control = NULL, ... ) ## S3 method for class 'formula' freqlist( formula, data, subset, na.action, na.options = c("include", "showexclude", "remove"), strata = NULL, labelTranslations = NULL, control = NULL, addNA, exclude, drop.unused.levels, ... )
object |
An R object, usually of class |
... |
additional arguments. In the formula method, these are passed to the table method. These are also passed to
|
na.options |
a character string indicating how to handling missing values: |
strata |
(formerly |
labelTranslations |
an optional character string (or list) of labels to use for variable levels when summarizing. Names will be matched appropriately. |
control |
control parameters to handle optional settings within |
formula, data, subset, na.action, addNA, exclude, drop.unused.levels
|
Arguments passed to |
An object of class c("freqlist", "arsenal_table")
Tina Gunderson, with revisions by Ethan Heinzen
arsenal_table, summary.freqlist, freq.control, freqlist.internal,
table, xtabs
# load mockstudy data data(mockstudy) tab.ex <- table(mockstudy[c("arm", "sex", "mdquality.s")], useNA = "ifany") noby <- freqlist(tab.ex, na.options = "include") summary(noby) # show the top 6 rows' frequencies and percents head(summary(sort(noby, decreasing = TRUE)[c(1:4, 6)])) withby <- freqlist(tab.ex, strata = c("arm","sex"), na.options = "showexclude") summary(withby)# load mockstudy data data(mockstudy) tab.ex <- table(mockstudy[c("arm", "sex", "mdquality.s")], useNA = "ifany") noby <- freqlist(tab.ex, na.options = "include") summary(noby) # show the top 6 rows' frequencies and percents head(summary(sort(noby, decreasing = TRUE)[c(1:4, 6)])) withby <- freqlist(tab.ex, strata = c("arm","sex"), na.options = "showexclude") summary(withby)
A set of helper functions for freqlist.
is.freqlist(x) is.summary.freqlist(x) ## S3 method for class 'summary.freqlist' head(x, n = 6L, ...) ## S3 method for class 'summary.freqlist' tail(x, n = 6L, ...) ## S3 method for class 'freqlist' sort(x, decreasing = FALSE, ...)is.freqlist(x) is.summary.freqlist(x) ## S3 method for class 'summary.freqlist' head(x, n = 6L, ...) ## S3 method for class 'summary.freqlist' tail(x, n = 6L, ...) ## S3 method for class 'freqlist' sort(x, decreasing = FALSE, ...)
x |
A |
n |
|
... |
Other arguments. |
decreasing |
Should the sort be increasing or decreasing? |
Note that sort() has to recalculate cumulative statistics. Note also that the reordering of rows
will also affect which labels are duplicates; you may also want to consider using
dupLabels=TRUE in freq.control().
merge.freqlist, arsenal_table, sort,
freqlist, summary.freqlist, freq.control,
Internal Functions
smart.split(string, width = Inf, min.split = -Inf) replace2(x, list, values)smart.split(string, width = Inf, min.split = -Inf) replace2(x, list, values)
string |
A character vector |
width |
Either |
min.split |
Either |
x |
vector |
list |
an index vector |
values |
replacement values |
For smart.split, a list of the same length as string, with each element being
the "intelligently" split string.
For replace2, a vector with the proper values replaced.
Keep the 'label' attribute on an R object when subsetting. loosen.labels allows the 'label'
attribute to be lost again.
keep.labels(x, ...) ## S3 method for class 'data.frame' keep.labels(x, ...) ## Default S3 method: keep.labels(x, ...) ## S3 method for class 'keep_labels' x[...] ## S3 replacement method for class 'keep_labels' x[i] <- value loosen.labels(x, ...) ## S3 method for class 'data.frame' loosen.labels(x, ...) ## Default S3 method: loosen.labels(x, ...)keep.labels(x, ...) ## S3 method for class 'data.frame' keep.labels(x, ...) ## Default S3 method: keep.labels(x, ...) ## S3 method for class 'keep_labels' x[...] ## S3 replacement method for class 'keep_labels' x[i] <- value loosen.labels(x, ...) ## S3 method for class 'data.frame' loosen.labels(x, ...) ## Default S3 method: loosen.labels(x, ...)
x |
An R object |
... |
Other arguments (not in use at this time). |
i, value
|
See |
A copy of x with a "keep_labels" class appended on or removed. Note that for the data.frame method,
only classes on the columns are changed; the data.frame won't have an extra class appended. This is different from previous
versions of arsenal.
Ethan Heinzen
Assign and extract the 'label' attribute on an R object. set_labels is
the same as labels(x) <- value but returns x for use in a pipe chain.
set_attr is the same as attr(x, which) <- value but returns x
for use in a pipe chain.
## S3 method for class 'data.frame' labels(object, ...) ## S3 method for class 'keep_labels' labels(object, ...) labels(x) <- value ## S3 replacement method for class 'keep_labels' labels(x) <- value ## Default S3 replacement method: labels(x) <- value ## S3 replacement method for class 'data.frame' labels(x) <- value set_labels(x, value) set_attr(x, which, value)## S3 method for class 'data.frame' labels(object, ...) ## S3 method for class 'keep_labels' labels(object, ...) labels(x) <- value ## S3 replacement method for class 'keep_labels' labels(x) <- value ## Default S3 replacement method: labels(x) <- value ## S3 replacement method for class 'data.frame' labels(x) <- value set_labels(x, value) set_attr(x, which, value)
... |
Other arguments (not in use at this time). |
x, object
|
An R object. |
value |
A vector or list containing labels to assign. Labels are assigned based on
names, if available; otherwise, they're assigned in order. Can pass |
which |
See |
The data.frame methods put labels on and extract labels from
the columns of object.
The labels of object, or object with new labels.
Ethan Heinzen
Convert numeric dates for month, day, and year to Date object, and vice versa.
mdy.Date(month, day, year, yearcut = 120) Date.mdy(date) is.Date(x)mdy.Date(month, day, year, yearcut = 120) Date.mdy(date) is.Date(x)
month |
integer, month (1-12). |
day |
integer, day of the month (1-31, depending on the month). |
year |
integer, either 2- or 4-digit year. If two-digit number, will add 1900 onto it, depending on range. |
yearcut |
cutoff for method to know if to convert to 4-digit year. |
date |
A date value. |
x |
An object. |
Test if an object is a date.
More work may need to be done with yearcut and 2-digit years. Best to give a full 4-digit year.
mdy.Date returns a Date object, and Date.mdy returns a list with integer values for month, day, and year.
is.Date returns a single logical value.
mdy.Date(9, 2, 2013) tmp <- mdy.Date(9, 2, 2013) Date.mdy(tmp) is.Date(tmp)mdy.Date(9, 2, 2013) tmp <- mdy.Date(9, 2, 2013) Date.mdy(tmp) is.Date(tmp)
Mock clinical study data for examples to test data manipulation and statistical functions.
The function muck_up_mockstudy() is used in examples for comparedf.
mockstudy muck_up_mockstudy()mockstudy muck_up_mockstudy()
A data frame with 1499 observations on the following 15 variables:
casea numeric identifier-patient ID
ageage in years
armtreatment arm divided into 3 groups, character string
sexa factor with levels Male Female
raceself-reported race/ethnicity, character string
fu.timesurvival or censoring time in years
fu.statcensoring status; 1=censor, 2=death
psinteger, ECOG performance score
hgbnumeric, hemoglobin count
bminumeric, body mass index, kg/m^2
alk.phosnumeric, alkaline phosphatase
astnumeric, aspartate transaminase
mdquality.sinteger, LASA QOL 0=Clinically Deficient, 1=Not Clinically Deficient
age.ordan ordered factor split of age, with levels
10-19 < 20-29 < 30-39 < 40-49 <
50-59 < 60-69 < 70-79 < 80-89
An object of class data.frame with 1499 rows and 14 columns.
data(mockstudy) str(mockstudy)data(mockstudy) str(mockstudy)
Fit and summarize models for each independent (x) variable with a response variable (y), with options to adjust by variables for each model.
modelsum( formula, family = "gaussian", data, adjust = NULL, na.action = NULL, subset = NULL, weights = NULL, id, strata, control = NULL, ... )modelsum( formula, family = "gaussian", data, adjust = NULL, na.action = NULL, subset = NULL, weights = NULL, id, strata, control = NULL, ... )
formula |
an object of class |
family |
similar mechanism to |
data |
an optional data.frame, list or environment (or object coercible by |
adjust |
an object of class |
na.action |
a function which indicates what should happen when the data contain |
subset |
an optional vector specifying a subset of observations (rows of |
weights |
an optional vector specifying the weights to apply to each data observation (rows of |
id |
A vector to identify clusters. Only used for |
strata |
a vector of strata to separate model summaries by an additional group. Note that for families like "clog",
the "usual" strata term to indicate subject groupings should be given in the |
control |
control parameters to handle optional settings within |
... |
additional arguments to be passed to internal |
An object with class c("modelsum", "arsenal_table")
Jason Sinnwell, Patrick Votruba, Beth Atkinson, Gregory Dougherty, and Ethan Heinzen, adapted from SAS Macro of the same name
arsenal_table, modelsum.control, summary.modelsum,
modelsum.internal, formulize
data(mockstudy) tab1 <- modelsum(bmi ~ sex + age, data = mockstudy) summary(tab1, text = TRUE) tab2 <- modelsum(alk.phos ~ arm + ps + hgb, adjust = ~ age + sex, family = "gaussian", data = mockstudy) summary(tab2, text = TRUE) summary(tab2, show.intercept = FALSE, text = TRUE) tab2.df <- as.data.frame(tab2) tab2.df[1:5,]data(mockstudy) tab1 <- modelsum(bmi ~ sex + age, data = mockstudy) summary(tab1, text = TRUE) tab2 <- modelsum(alk.phos ~ arm + ps + hgb, adjust = ~ age + sex, family = "gaussian", data = mockstudy) summary(tab2, text = TRUE) summary(tab2, show.intercept = FALSE, text = TRUE) tab2.df <- as.data.frame(tab2) tab2.df[1:5,]
modelsum functionControl test and summary settings for modelsum function.
modelsum.control( digits = 3L, digits.ratio = 3L, digits.p = 3L, format.p = TRUE, show.adjust = TRUE, show.intercept = TRUE, conf.level = 0.95, ordinal.stats = c("OR", "CI.lower.OR", "CI.upper.OR", "p.value", "Nmiss"), binomial.stats = c("OR", "CI.lower.OR", "CI.upper.OR", "p.value", "concordance", "Nmiss"), gaussian.stats = c("estimate", "std.error", "p.value", "adj.r.squared", "Nmiss"), poisson.stats = c("RR", "CI.lower.RR", "CI.upper.RR", "p.value", "Nmiss"), negbin.stats = c("RR", "CI.lower.RR", "CI.upper.RR", "p.value", "Nmiss"), relrisk.stats = c("RR", "CI.lower.RR", "CI.upper.RR", "p.value", "Nmiss"), clog.stats = c("OR", "CI.lower.OR", "CI.upper.OR", "p.value", "concordance", "Nmiss"), survival.stats = c("HR", "CI.lower.HR", "CI.upper.HR", "p.value", "concordance", "Nmiss"), stat.labels = list(), ... )modelsum.control( digits = 3L, digits.ratio = 3L, digits.p = 3L, format.p = TRUE, show.adjust = TRUE, show.intercept = TRUE, conf.level = 0.95, ordinal.stats = c("OR", "CI.lower.OR", "CI.upper.OR", "p.value", "Nmiss"), binomial.stats = c("OR", "CI.lower.OR", "CI.upper.OR", "p.value", "concordance", "Nmiss"), gaussian.stats = c("estimate", "std.error", "p.value", "adj.r.squared", "Nmiss"), poisson.stats = c("RR", "CI.lower.RR", "CI.upper.RR", "p.value", "Nmiss"), negbin.stats = c("RR", "CI.lower.RR", "CI.upper.RR", "p.value", "Nmiss"), relrisk.stats = c("RR", "CI.lower.RR", "CI.upper.RR", "p.value", "Nmiss"), clog.stats = c("OR", "CI.lower.OR", "CI.upper.OR", "p.value", "concordance", "Nmiss"), survival.stats = c("HR", "CI.lower.HR", "CI.upper.HR", "p.value", "concordance", "Nmiss"), stat.labels = list(), ... )
digits |
Numeric, denoting the number of digits after the decimal point for beta coefficients and standard errors. |
digits.ratio |
Numeric, denoting the number of digits after the decimal point for ratios, e.g. OR, RR, HR. |
digits.p |
Numeric, denoting the number of digits for p-values. See "Details", below. |
format.p |
Logical, denoting whether to format p-values. See "Details", below. |
show.adjust |
Logical, denoting whether to show adjustment terms. |
show.intercept |
Logical, denoting whether to show intercept terms. |
conf.level |
Numeric, giving the confidence level. |
ordinal.stats, binomial.stats, survival.stats, gaussian.stats, poisson.stats, negbin.stats, clog.stats, relrisk.stats
|
Character vectors denoting which stats to show for the various model types. |
stat.labels |
A named list of labels for all the stats used above. |
... |
Other arguments (not in use at this time). |
If format.p is FALSE, digits.p denotes the number of significant digits shown. The
p-values will be in exponential notation if necessary. If format.p is TRUE,
digits.p will determine the number of digits after the decimal point to show. If the p-value
is less than the resulting number of places, it will be formatted to show so.
A list with settings to be used within the modelsum function.
modelsum, summary.modelsum, modelsum.internal
A set of family functions for modelsum.
survival() ordinal(method = c("logistic", "probit", "loglog", "cloglog", "cauchit")) negbin(link = c("log", "identity", "sqrt")) clog() relrisk(link = "log")survival() ordinal(method = c("logistic", "probit", "loglog", "cloglog", "cauchit")) negbin(link = c("log", "identity", "sqrt")) clog() relrisk(link = "log")
method |
See |
link |
See |
A list, in particular with element family.
A set of helper functions for modelsum.
is.modelsum(x) is.summary.modelsum(x) na.modelsum(object, ...)is.modelsum(x) is.summary.modelsum(x) na.modelsum(object, ...)
x |
A |
object |
A |
... |
Other arguments, or a vector of indices for extracting. |
na.modelsum returns a subsetted version of object (with attributes).
allNA tests if all elements are NA, and includeNA sets the
NAs in a character vector or factor to an explicit label.
allNA(x) includeNA(x, label, ...) ## S3 method for class 'factor' includeNA(x, label = "(Missing)", first = FALSE, ...) ## Default S3 method: includeNA(x, label = "(Missing)", ...)allNA(x) includeNA(x, label, ...) ## S3 method for class 'factor' includeNA(x, label = "(Missing)", first = FALSE, ...) ## Default S3 method: includeNA(x, label = "(Missing)", ...)
x |
An object |
label |
A character string denoting the label to set |
... |
Other arguments (not in use at this time). |
first |
Logical; should the new label be the first level? |
Ethan Heinzen
Adjust P-values for Multiple Comparisons
padjust(p, method, n, ...) ## Default S3 method: padjust(p, method, n, ...) ## S3 method for class 'tableby' padjust(p, method, n, suffix = " (adjusted for multiple comparisons)", ...) ## S3 method for class 'summary.tableby' padjust(p, method, n, suffix = " (adjusted for multiple comparisons)", ...)padjust(p, method, n, ...) ## Default S3 method: padjust(p, method, n, ...) ## S3 method for class 'tableby' padjust(p, method, n, suffix = " (adjusted for multiple comparisons)", ...) ## S3 method for class 'summary.tableby' padjust(p, method, n, suffix = " (adjusted for multiple comparisons)", ...)
p |
An object. |
method |
correction method. Can be abbreviated. |
n |
number of comparisons, must be at least |
... |
Other arguments. |
suffix |
A suffix to add to the footnotes indicating that the tests were adjusted. |
p.adjust, modpval.tableby, tests.tableby
Summarize one or more variables (x) by a paired time variable (y). Variables on the right side of the formula, i.e. independent variables, are summarized by the two time points on the left of the formula. Optionally, an appropriate test is performed to test the distribution of the independent variables across the time points.
paired( formula, data, id, na.action, subset = NULL, strata, control = NULL, ... )paired( formula, data, id, na.action, subset = NULL, strata, control = NULL, ... )
formula |
an object of class |
data |
an optional data frame, list or environment (or object coercible by |
id |
The vector giving IDs to match up data for the same subject across two timepoints. |
na.action |
a function which indicates what should happen when the data contain |
subset |
an optional vector specifying a subset of observations (rows of data) to be used in the results. Works as vector of logicals or an index. |
strata |
a vector of strata to separate summaries by an additional group. |
control |
control parameters to handle optional settings within |
... |
additional arguments to be passed to internal |
Do note that this function piggybacks off of tableby quite heavily, so there is no
summary.paired function (for instance).
These tests are accepted:
paired.t: a paired t-test.
mcnemar: McNemar's test.
signed.rank: a signed rank test.
sign.test: a sign test.
notest: no test is performed.
An object with class c("paired", "tableby", "arsenal_table")
Jason Sinnwell, Beth Atkinson, Ryan Lennon, and Ethan Heinzen
arsenal_table, paired.control, tableby, formulize, selectall
paired functionControl test and summary settings for the paired function.
paired.control( diff = TRUE, numeric.test = "paired.t", cat.test = "mcnemar", ordered.test = "signed.rank", date.test = "paired.t", mcnemar.correct = TRUE, signed.rank.exact = NULL, signed.rank.correct = TRUE, ... )paired.control( diff = TRUE, numeric.test = "paired.t", cat.test = "mcnemar", ordered.test = "signed.rank", date.test = "paired.t", mcnemar.correct = TRUE, signed.rank.exact = NULL, signed.rank.correct = TRUE, ... )
diff |
logical, telling |
numeric.test |
name of test for numeric RHS variables in |
cat.test |
name of test for categorical variables: mcnemar |
ordered.test |
name of test for ordered variables: signed.rank, sign.test |
date.test |
name of test to perform for date variables: paired.t, signed.rank, sign.test |
mcnemar.correct, signed.rank.exact, signed.rank.correct
|
Options for statistical tests. See |
... |
Arguments passed to |
Note that (with the exception of total) all arguments to tableby.control are accepted in
this function (in fact, this function passes everything through to tableby.control).
However, there are different defaults for the statistical tests (shown here). For details on the other arguments,
please see the help page for tableby.control.
A list with settings to be used within the paired function.
Ethan Heinzen
paired, tableby, tableby.control, summary.tableby
A set of helper functions for paired.
na.paired(missings = c("in.both", "fill", "asis"))na.paired(missings = c("in.both", "fill", "asis"))
missings |
A character string denoting which action to take. See "Details", below. |
All methods subset out any NA time points or IDs.
"in.both" (the default) subsets the data.frame to individuals who appear at both time points.
"fill" adds explicit missings for the people missing second time points.
"asis" does nothing to add or remove missings.
na.paired returns a function used to subset data.frames in paired.
Make a column for "select all" input
selectall(...) as.selectall(x) ## S3 method for class 'selectall' as.matrix(x, ...) ## S3 method for class 'selectall' x[i, j, drop = FALSE] ## S3 method for class 'selectall' is.na(x) is.selectall(x)selectall(...) as.selectall(x) ## S3 method for class 'selectall' as.matrix(x, ...) ## S3 method for class 'selectall' x[i, j, drop = FALSE] ## S3 method for class 'selectall' is.na(x) is.selectall(x)
... |
Named arguments of the same length. These should be logical, numeric (0/1) or a factor with two levels. |
x |
An object of class "selectall" |
i, j, drop
|
Arguments to '[.matrix' |
d <- data.frame(grp = rep(c("A", "B"), each = 5)) d$s <- selectall( `Option 1` = c(rep(1, 4), rep(0, 6)), `Option 2` = c(0, 1, 0, 0, 0, 1, 1, 1, 0, 0), `Option 3` = 1, `Option 4` = 0 ) summary(tableby(grp ~ s, data = d), text = TRUE)d <- data.frame(grp = rep(c("A", "B"), each = 5)) d$s <- selectall( `Option 1` = c(rep(1, 4), rep(0, 6)), `Option 2` = c(0, 1, 0, 0, 0, 1, 1, 1, 0, 0), `Option 3` = 1, `Option 4` = 0 ) summary(tableby(grp ~ s, data = d), text = TRUE)
comparedf objectPrint a more detailed output of the comparedf object.
## S3 method for class 'comparedf' summary(object, ..., show.attrs = FALSE) ## S3 method for class 'summary.comparedf' print(x, ..., format = "pandoc")## S3 method for class 'comparedf' summary(object, ..., show.attrs = FALSE) ## S3 method for class 'summary.comparedf' print(x, ..., format = "pandoc")
object |
An object of class |
... |
Other arguments passed to |
show.attrs |
Logical, denoting whether to show the actual attributes which are different. For (e.g.) factors with lots
of levels, this can make the tables quite wide, so this feature is |
x |
An object returned by the |
format |
Passed to |
An object of class "summary.comparedf" is returned.
Summarize the freqlist object.
## S3 method for class 'freqlist' summary(object, ..., labelTranslations = NULL, title = NULL) ## S3 method for class 'summary.freqlist' as.data.frame(x, ..., list.ok = FALSE)## S3 method for class 'freqlist' summary(object, ..., labelTranslations = NULL, title = NULL) ## S3 method for class 'summary.freqlist' as.data.frame(x, ..., list.ok = FALSE)
object |
an object of class |
... |
For |
labelTranslations |
A named list (or vector) where the name is the label in the
output to be replaced in the pretty rendering by the character string
value for the named element of the list, e.g., |
title |
Title/caption for the table, defaulting to |
x |
An object of class |
list.ok |
If the object has multiple by-variables, is it okay to return a list of data.frames instead of a single data.frame?
If |
An object of class "summary.freqlist" (invisibly for the print method).
Tina Gunderson, with major revisions by Ethan Heinzen
freqlist, table, xtabs, kable,
freq.control, freqlist.internal
# load mockstudy data data(mockstudy) tab.ex <- table(mockstudy[c("arm", "sex", "mdquality.s")], useNA = "ifany") noby <- freqlist(tab.ex, na.options = "include") summary(noby) withby <- freqlist(tab.ex, strata = c("arm","sex"), na.options = "showexclude") summary(withby) summary(withby, dupLabels = TRUE)# load mockstudy data data(mockstudy) tab.ex <- table(mockstudy[c("arm", "sex", "mdquality.s")], useNA = "ifany") noby <- freqlist(tab.ex, na.options = "include") summary(noby) withby <- freqlist(tab.ex, strata = c("arm","sex"), na.options = "showexclude") summary(withby) summary(withby, dupLabels = TRUE)
modelsum object.Format the information in object as a table using Pandoc coding or plain text, and cat it to stdout.
## S3 method for class 'modelsum' summary( object, ..., labelTranslations = NULL, text = FALSE, title = NULL, term.name = "", adjustment.names = FALSE ) ## S3 method for class 'summary.modelsum' as.data.frame( x, ..., text = x$text, term.name = x$term.name, adjustment.names = x$adjustment.names, width = NULL, min.split = NULL, list.ok = FALSE )## S3 method for class 'modelsum' summary( object, ..., labelTranslations = NULL, text = FALSE, title = NULL, term.name = "", adjustment.names = FALSE ) ## S3 method for class 'summary.modelsum' as.data.frame( x, ..., text = x$text, term.name = x$term.name, adjustment.names = x$adjustment.names, width = NULL, min.split = NULL, list.ok = FALSE )
object |
A |
... |
For |
labelTranslations |
A named list (or vector) where the name is the label in the
output to be replaced in the pretty rendering by the character string
value for the named element of the list, e.g., |
text |
An argument denoting how to print the summary to the screen.
Default is |
title |
Title/caption for the table, defaulting to |
term.name |
A character vector denoting the column name for the "terms" column. It should be the same length
as the number of tables or less (it will get recycled if needed). The special value |
adjustment.names |
Logical, denoting whether the names of the adjustment models should be printed. |
x |
An object of class |
width |
Passed to |
min.split |
Passed to |
list.ok |
If the object has multiple by-variables, is it okay to return a list of data.frames instead of a single data.frame?
If |
An object of class "summary.modelsum"
Ethan Heinzen, based on code originally by Greg Dougherty
modelsum, as.data.frame.modelsum
tableby objectThe summary method for a tableby object, which is a pretty rendering of a tableby
object into a publication-quality results table in R Markdown, and can render well in text-only.
## S3 method for class 'tableby' summary( object, ..., labelTranslations = NULL, text = FALSE, title = NULL, pfootnote = FALSE, term.name = "" ) ## S3 method for class 'summary.tableby' as.data.frame( x, ..., text = x$text, pfootnote = x$pfootnote, term.name = x$term.name, width = NULL, min.split = NULL, list.ok = FALSE )## S3 method for class 'tableby' summary( object, ..., labelTranslations = NULL, text = FALSE, title = NULL, pfootnote = FALSE, term.name = "" ) ## S3 method for class 'summary.tableby' as.data.frame( x, ..., text = x$text, pfootnote = x$pfootnote, term.name = x$term.name, width = NULL, min.split = NULL, list.ok = FALSE )
object |
An object of class |
... |
For |
labelTranslations |
A named list (or vector) where the name is the label in the
output to be replaced in the pretty rendering by the character string
value for the named element of the list, e.g., |
text |
An argument denoting how to print the summary to the screen.
Default is |
title |
Title/caption for the table, defaulting to |
pfootnote |
Logical, denoting whether to put footnotes describing the tests used to generate the p-values. Alternatively,
"html" to surround the outputted footnotes with |
term.name |
A character vector denoting the column name for the "terms" column. It should be the same length
as the number of tables or less (it will get recycled if needed). The special value |
x |
An object of class |
width |
Passed to |
min.split |
Passed to |
list.ok |
If the object has multiple by-variables, is it okay to return a list of data.frames instead of a single data.frame?
If |
An object of class summary.tableby
Ethan Heinzen, based on code by Gregory Dougherty, Jason Sinnwell, Beth Atkinson, adapted from SAS Macros written by Paul Novotny and Ryan Lennon
set.seed(100) ## make 3+ categories for response nsubj <- 90 mdat <- data.frame(Response=sample(c(1,2,3),nsubj, replace=TRUE), Sex=sample(c("Male", "Female"), nsubj,replace=TRUE), Age=round(rnorm(nsubj,mean=40, sd=5)), HtIn=round(rnorm(nsubj,mean=65,sd=5))) ## allow default summaries on RHS variables out <- tableby(Response ~ Sex + Age + HtIn, data=mdat) summary(out, text=TRUE) labels(out) labels(out) <- c(Age="Age (years)", HtIn="Height (inches)") summary(out, stats.labels=c(meansd="Mean-SD", q1q3 = "Q1-Q3"), text=TRUE)set.seed(100) ## make 3+ categories for response nsubj <- 90 mdat <- data.frame(Response=sample(c(1,2,3),nsubj, replace=TRUE), Sex=sample(c("Male", "Female"), nsubj,replace=TRUE), Age=round(rnorm(nsubj,mean=40, sd=5)), HtIn=round(rnorm(nsubj,mean=65,sd=5))) ## allow default summaries on RHS variables out <- tableby(Response ~ Sex + Age + HtIn, data=mdat) summary(out, text=TRUE) labels(out) labels(out) <- c(Age="Age (years)", HtIn="Height (inches)") summary(out, stats.labels=c(meansd="Mean-SD", q1q3 = "Q1-Q3"), text=TRUE)
Summarize one or more variables (x) by a categorical variable (y). Variables on the right side of the formula, i.e. independent variables, are summarized by the levels of a categorical variable on the left of the formula. Optionally, an appropriate test is performed to test the distribution of the independent variables across the levels of the categorical variable.
tableby( formula, data, na.action, subset = NULL, weights = NULL, strata, control = NULL, ... )tableby( formula, data, na.action, subset = NULL, weights = NULL, strata, control = NULL, ... )
formula |
an object of class |
data |
an optional data frame, list or environment (or object coercible by |
na.action |
a function which indicates what should happen when the data contain |
subset |
an optional vector specifying a subset of observations (rows of data) to be used in the results. Works as vector of logicals or an index. |
weights |
a vector of weights. Using weights will disable statistical tests. |
strata |
a vector of strata to separate summaries by an additional group. |
control |
control parameters to handle optional settings within |
... |
additional arguments to be passed to internal |
The group variable (if any) is categorical, which could be an integer, character,
factor, or ordered factor. tableby makes a simple summary of
the counts within the k-levels of the independent variables on the
right side of the formula. Note that unused levels are dropped.
The data argument allows data.frames with label attributes for the columns, and those
labels will be used in the summary methods for the tableby class.
The independent variables are a mixture of types: categorical (discrete), numeric (continuous), and time to event (survival). These variables are split by the levels of the group variable (if any), then summarized within those levels, specific to the variable type. A statistical test is performed to compare the distribution of the independent variables across the levels of the grouping variable.
The tests differ by the independent variable type, but can be specified explicitly in the formula statement or in the control function. These tests are accepted:
anova: analysis of variance test; the default test for continuous variables. When
LHS variable has two levels, equivalent to two-sample t-test.
kwt: Kruskal-Wallis Rank Test, optional test for continuous
variables. When LHS variable has two levels, equivalent to Wilcoxon test.
wt: An explicit Wilcoxon test.
medtest: A median test.
chisq: chi-square goodness of fit test for equal counts of a
categorical variable across categories; the default for categorical
or factor variables
fe: Fisher's exact test for categorical variables
trend: trend test for equal distribution of an ordered variable
across a categorical variable; the default for ordered factor variables
logrank: log-rank, the default for time-to-event variables
notest: no test is performed.
To perform a mixture of asymptotic and rank-based tests on two
different continuous variables, an example formula is:
formula = group ~ anova(age) + kwt(height). The test settings
in tableby.control apply to all independent variables of a given type.
The summary statistics reported for each independent variable within the
group variable can be set in tableby.control.
Finally, multiple by-variables can be set using list(). See the examples for more details.
An object with class c("tableby", "arsenal_table")
Jason Sinnwell, Beth Atkinson, Gregory Dougherty, and Ethan Heinzen, adapted from SAS Macros written by Paul Novotny and Ryan Lennon
arsenal_table, anova, chisq.test, tableby.control,
summary.tableby, tableby.internal, formulize, selectall
data(mockstudy) tab1 <- tableby(arm ~ sex + age, data=mockstudy) summary(tab1, text=TRUE) mylabels <- list(sex = "SEX", age ="Age, yrs") summary(tab1, labelTranslations = mylabels, text=TRUE) tab3 <- tableby(arm ~ sex + age, data=mockstudy, test=FALSE, total=FALSE, numeric.stats=c("median","q1q3"), numeric.test="kwt") summary(tab3, text=TRUE) # multiple LHS summary(tableby(list(arm, sex) ~ age, data = mockstudy, strata = ps), text = TRUE) tab.test <- tableby(arm ~ kwt(age) + anova(bmi) + kwt(ast), data=mockstudy) tests(tab.test)data(mockstudy) tab1 <- tableby(arm ~ sex + age, data=mockstudy) summary(tab1, text=TRUE) mylabels <- list(sex = "SEX", age ="Age, yrs") summary(tab1, labelTranslations = mylabels, text=TRUE) tab3 <- tableby(arm ~ sex + age, data=mockstudy, test=FALSE, total=FALSE, numeric.stats=c("median","q1q3"), numeric.test="kwt") summary(tab3, text=TRUE) # multiple LHS summary(tableby(list(arm, sex) ~ age, data = mockstudy, strata = ps), text = TRUE) tab.test <- tableby(arm ~ kwt(age) + anova(bmi) + kwt(ast), data=mockstudy) tests(tab.test)
tableby functionControl test and summary settings for the tableby function.
tableby.control( test = TRUE, total = TRUE, total.pos = c("after", "before"), test.pname = NULL, numeric.simplify = FALSE, cat.simplify = FALSE, cat.droplevels = FALSE, ordered.simplify = FALSE, date.simplify = FALSE, numeric.test = "anova", cat.test = "chisq", ordered.test = "trend", surv.test = "logrank", date.test = "kwt", selectall.test = "notest", test.always = FALSE, numeric.stats = c("Nmiss", "meansd", "range"), cat.stats = c("Nmiss", "countpct"), ordered.stats = c("Nmiss", "countpct"), surv.stats = c("Nmiss", "Nevents", "medSurv"), date.stats = c("Nmiss", "median", "range"), selectall.stats = c("Nmiss", "countpct"), stats.labels = list(), digits = 3L, digits.count = 0L, digits.pct = 1L, digits.p = 3L, format.p = TRUE, digits.n = 0L, conf.level = 0.95, wilcox.correct = FALSE, wilcox.exact = NULL, chisq.correct = FALSE, simulate.p.value = FALSE, B = 2000, times = 1:5, ... )tableby.control( test = TRUE, total = TRUE, total.pos = c("after", "before"), test.pname = NULL, numeric.simplify = FALSE, cat.simplify = FALSE, cat.droplevels = FALSE, ordered.simplify = FALSE, date.simplify = FALSE, numeric.test = "anova", cat.test = "chisq", ordered.test = "trend", surv.test = "logrank", date.test = "kwt", selectall.test = "notest", test.always = FALSE, numeric.stats = c("Nmiss", "meansd", "range"), cat.stats = c("Nmiss", "countpct"), ordered.stats = c("Nmiss", "countpct"), surv.stats = c("Nmiss", "Nevents", "medSurv"), date.stats = c("Nmiss", "median", "range"), selectall.stats = c("Nmiss", "countpct"), stats.labels = list(), digits = 3L, digits.count = 0L, digits.pct = 1L, digits.p = 3L, format.p = TRUE, digits.n = 0L, conf.level = 0.95, wilcox.correct = FALSE, wilcox.exact = NULL, chisq.correct = FALSE, simulate.p.value = FALSE, B = 2000, times = 1:5, ... )
test |
logical, telling |
total |
logical, telling |
total.pos |
One of |
test.pname |
character string denoting the p-value column name in |
numeric.simplify, date.simplify
|
logical, tell |
cat.simplify, ordered.simplify
|
logical, tell |
cat.droplevels |
Should levels be dropped for categorical variables? If set to true, p-values will not be displayed
unless |
numeric.test |
name of test for numeric RHS variables in |
cat.test |
name of test for categorical variables: chisq, fe (Fisher's Exact) |
ordered.test |
name of test for ordered variables: trend |
surv.test |
name of test for survival variables: logrank |
date.test |
name of test for date variables: kwt |
selectall.test |
name of test for date variables: notest |
test.always |
Should the test be performed even if one or more by-group has 0 observations? Relevant for kwt and anova. |
numeric.stats, cat.stats, ordered.stats, surv.stats, date.stats, selectall.stats
|
summary statistics to include for the respective class of RHS variables within the levels of the group LHS variable. |
stats.labels |
A named list of labels for all the statistics function names, where the function name is the named element in the list
and the value that goes with it is a string containing the formal name that will be printed in all printed renderings of the output,
e.g., |
digits |
Number of decimal places for numeric values. |
digits.count |
Number of decimal places for count values. |
digits.pct |
Number of decimal places for percents. |
digits.p |
Number of decimal places for p-values. |
format.p |
Logical, denoting whether to format p-values, or character, a |
digits.n |
Number of decimal places for N's in the header. Set it to NA to suppress the N's. |
conf.level |
Numeric, denoting what confidence level to use for confidence intervals.
(See, e.g., |
wilcox.correct, wilcox.exact
|
See |
chisq.correct |
logical, correction factor for chisq.test |
simulate.p.value |
logical, simulate p-value for categorical tests (fe and chisq) |
B |
number of simulations to perform for simulation-based p-value |
times |
A vector of times to use for survival summaries. |
... |
additional arguments. |
All tests can be turned off by setting test to FALSE.
Otherwise, test are set to default settings in this list, or set explicitly in the formula of tableby.
If format.p is FALSE, digits.p denotes the number of significant digits shown. The
p-values will be in exponential notation if necessary. If format.p is TRUE,
digits.p will determine the number of digits after the decimal point to show. If the p-value
is less than the resulting number of places, it will be formatted to show so. If format.p is a character string,
it will be treated as a glue specification: the p-value is exposed as "p", and "digits.p" as "digits.p".
Options for statistics are described more thoroughly in the vignette and are listed in tableby.stats
A list with settings to be used within the tableby function.
Jason Sinnwell, Beth Atkinson, Ethan Heinzen, Terry Therneau, adapted from SAS Macros written by Paul Novotny and Ryan Lennon
anova, chisq.test, tableby, summary.tableby,
tableby.stats.
set.seed(100) ## make 3+ categories for Response mdat <- data.frame(Response=c(0,0,0,0,0,1,1,1,1,1), Sex=sample(c("Male", "Female"), 10,replace=TRUE), Age=round(rnorm(10,mean=40, sd=5)), HtIn=round(rnorm(10,mean=65,sd=5))) ## allow default summaries in RHS variables, and pass control args to ## main function, to be picked up with ... when calling tableby.control outResp <- tableby(Response ~ Sex + Age + HtIn, data=mdat, total=FALSE, test=TRUE) outCtl <- tableby(Response ~ Sex + Age + HtIn, data=mdat, control=tableby.control(total=TRUE, cat.simplify=TRUE, cat.stats=c("Nmiss","countpct"),digits=1)) summary(outResp, text=TRUE) summary(outCtl, text=TRUE)set.seed(100) ## make 3+ categories for Response mdat <- data.frame(Response=c(0,0,0,0,0,1,1,1,1,1), Sex=sample(c("Male", "Female"), 10,replace=TRUE), Age=round(rnorm(10,mean=40, sd=5)), HtIn=round(rnorm(10,mean=65,sd=5))) ## allow default summaries in RHS variables, and pass control args to ## main function, to be picked up with ... when calling tableby.control outResp <- tableby(Response ~ Sex + Age + HtIn, data=mdat, total=FALSE, test=TRUE) outCtl <- tableby(Response ~ Sex + Age + HtIn, data=mdat, control=tableby.control(total=TRUE, cat.simplify=TRUE, cat.stats=c("Nmiss","countpct"),digits=1)) summary(outResp, text=TRUE) summary(outCtl, text=TRUE)
A set of helper functions for tableby.
is.tableby(x) is.summary.tableby(x) modpval.tableby(x, pdata, use.pname = FALSE) tests(x) ## S3 method for class 'tableby' tests(x) na.tableby(lhs = TRUE) ## S3 method for class 'tableby' xtfrm(x) ## S3 method for class 'tableby' sort(x, ...) ## S3 method for class 'tableby' Ops(e1, e2) ## S3 method for class 'tableby' head(x, n = 6L, ...) ## S3 method for class 'tableby' tail(x, n = 6L, ...)is.tableby(x) is.summary.tableby(x) modpval.tableby(x, pdata, use.pname = FALSE) tests(x) ## S3 method for class 'tableby' tests(x) na.tableby(lhs = TRUE) ## S3 method for class 'tableby' xtfrm(x) ## S3 method for class 'tableby' sort(x, ...) ## S3 method for class 'tableby' Ops(e1, e2) ## S3 method for class 'tableby' head(x, n = 6L, ...) ## S3 method for class 'tableby' tail(x, n = 6L, ...)
x |
A |
pdata |
A named data.frame where the first column is the by-variable names, the (optional) second is the strata value, the next is the x variable names, the next is p-values (or some test stat), and the (optional) next column is the method name. |
use.pname |
Logical, denoting whether the column name in |
lhs |
Logical, denoting whether to remove |
... |
Other arguments. |
e1, e2
|
|
n |
Logical comparisons are implemented for Ops.tableby.
na.tableby returns a subsetted version of object (with attributes). Ops.tableby returns
a logical vector. xtfrm.tableby returns the p-values (which are ordered by order to sort).
arsenal_table, sort, head, tail,
tableby, summary.tableby, tableby.control
A collection of functions that will report summary statistics. To create a custom function,
consider using a function with all three arguments and .... See the tableby vignette
for an example.
arsenal_sum(x, na.rm = TRUE, ...) arsenal_min(x, na.rm = TRUE, ...) arsenal_max(x, na.rm = TRUE, ...) arsenal_mean(x, na.rm = TRUE, weights = NULL, ...) arsenal_sd(x, na.rm = TRUE, weights = NULL, ...) arsenal_var(x, na.rm = TRUE, weights = NULL, ...) meansd(x, na.rm = TRUE, weights = NULL, ...) meanse(x, na.rm = TRUE, weights = NULL, ...) meanpmsd(x, na.rm = TRUE, weights = NULL, ...) meanpmse(x, na.rm = TRUE, weights = NULL, ...) meanCI(x, na.rm = TRUE, weights = NULL, conf.level = 0.95, ...) medianrange(x, na.rm = TRUE, weights = NULL, ...) medianmad(x, na.rm = TRUE, weights = NULL, ...) arsenal_median(x, na.rm = TRUE, weights = NULL, ...) arsenal_range(x, na.rm = TRUE, ...) gmean(x, na.rm = TRUE, weights = NULL, ...) gsd(x, na.rm = TRUE, weights = NULL, ...) gmeansd(x, na.rm = TRUE, weights = NULL, ...) gmeanCI(x, na.rm = TRUE, weights = NULL, conf.level = 0.95, ...) Nsigntest(x, na.rm = TRUE, weights = NULL, ...) Nevents(x, na.rm = TRUE, weights = NULL, ...) medSurv(x, na.rm = TRUE, weights = NULL, ...) NeventsSurv(x, na.rm = TRUE, weights = NULL, times = 1:5, ...) NriskSurv(x, na.rm = TRUE, weights = NULL, times = 1:5, ...) Nrisk(x, na.rm = TRUE, weights = NULL, times = 1:5, ...) medTime(x, na.rm = TRUE, weights = NULL, ...) q1q3(x, na.rm = TRUE, weights = NULL, ...) medianq1q3(x, na.rm = TRUE, weights = NULL, ...) iqr(x, na.rm = TRUE, weights = NULL, ...) Nmiss(x, weights = NULL, ...) Nmisspct(x, weights = NULL, ...) Nmiss2(x, weights = NULL, ...) Nmisspct2(x, weights = NULL, ...) N(x, weights = NULL, ...) Npct(x, weights = NULL, ...) Nrowpct( x, levels = NULL, by, by.levels = sort(unique(by)), weights = NULL, ..., totallab = "Total" ) count(x, levels = NULL, na.rm = TRUE, weights = NULL, ...) countpct(x, levels = NULL, na.rm = TRUE, weights = NULL, ...) pct(x, levels = NULL, na.rm = TRUE, weights = NULL, ...) countN(x, levels = NULL, na.rm = TRUE, weights = NULL, ...) countrowpct( x, levels = NULL, by, by.levels = sort(unique(by)), na.rm = TRUE, weights = NULL, ..., totallab = "Total" ) rowpct( x, levels = NULL, by, by.levels = sort(unique(by)), na.rm = TRUE, weights = NULL, ..., totallab = "Total" ) countcellpct( x, levels = NULL, by, by.levels = sort(unique(by)), na.rm = TRUE, weights = NULL, ..., totallab = "Total" ) binomCI(x, levels = NULL, na.rm = TRUE, weights = NULL, conf.level = 0.95, ...) rowbinomCI( x, levels = NULL, by, by.levels = sort(unique(by)), na.rm = TRUE, weights = NULL, conf.level = 0.95, ..., totallab = "Total" )arsenal_sum(x, na.rm = TRUE, ...) arsenal_min(x, na.rm = TRUE, ...) arsenal_max(x, na.rm = TRUE, ...) arsenal_mean(x, na.rm = TRUE, weights = NULL, ...) arsenal_sd(x, na.rm = TRUE, weights = NULL, ...) arsenal_var(x, na.rm = TRUE, weights = NULL, ...) meansd(x, na.rm = TRUE, weights = NULL, ...) meanse(x, na.rm = TRUE, weights = NULL, ...) meanpmsd(x, na.rm = TRUE, weights = NULL, ...) meanpmse(x, na.rm = TRUE, weights = NULL, ...) meanCI(x, na.rm = TRUE, weights = NULL, conf.level = 0.95, ...) medianrange(x, na.rm = TRUE, weights = NULL, ...) medianmad(x, na.rm = TRUE, weights = NULL, ...) arsenal_median(x, na.rm = TRUE, weights = NULL, ...) arsenal_range(x, na.rm = TRUE, ...) gmean(x, na.rm = TRUE, weights = NULL, ...) gsd(x, na.rm = TRUE, weights = NULL, ...) gmeansd(x, na.rm = TRUE, weights = NULL, ...) gmeanCI(x, na.rm = TRUE, weights = NULL, conf.level = 0.95, ...) Nsigntest(x, na.rm = TRUE, weights = NULL, ...) Nevents(x, na.rm = TRUE, weights = NULL, ...) medSurv(x, na.rm = TRUE, weights = NULL, ...) NeventsSurv(x, na.rm = TRUE, weights = NULL, times = 1:5, ...) NriskSurv(x, na.rm = TRUE, weights = NULL, times = 1:5, ...) Nrisk(x, na.rm = TRUE, weights = NULL, times = 1:5, ...) medTime(x, na.rm = TRUE, weights = NULL, ...) q1q3(x, na.rm = TRUE, weights = NULL, ...) medianq1q3(x, na.rm = TRUE, weights = NULL, ...) iqr(x, na.rm = TRUE, weights = NULL, ...) Nmiss(x, weights = NULL, ...) Nmisspct(x, weights = NULL, ...) Nmiss2(x, weights = NULL, ...) Nmisspct2(x, weights = NULL, ...) N(x, weights = NULL, ...) Npct(x, weights = NULL, ...) Nrowpct( x, levels = NULL, by, by.levels = sort(unique(by)), weights = NULL, ..., totallab = "Total" ) count(x, levels = NULL, na.rm = TRUE, weights = NULL, ...) countpct(x, levels = NULL, na.rm = TRUE, weights = NULL, ...) pct(x, levels = NULL, na.rm = TRUE, weights = NULL, ...) countN(x, levels = NULL, na.rm = TRUE, weights = NULL, ...) countrowpct( x, levels = NULL, by, by.levels = sort(unique(by)), na.rm = TRUE, weights = NULL, ..., totallab = "Total" ) rowpct( x, levels = NULL, by, by.levels = sort(unique(by)), na.rm = TRUE, weights = NULL, ..., totallab = "Total" ) countcellpct( x, levels = NULL, by, by.levels = sort(unique(by)), na.rm = TRUE, weights = NULL, ..., totallab = "Total" ) binomCI(x, levels = NULL, na.rm = TRUE, weights = NULL, conf.level = 0.95, ...) rowbinomCI( x, levels = NULL, by, by.levels = sort(unique(by)), na.rm = TRUE, weights = NULL, conf.level = 0.95, ..., totallab = "Total" )
x |
Usually a vector. |
na.rm |
Should NAs be removed? |
... |
Other arguments. |
weights |
A vector of weights. |
conf.level |
Numeric, denoting what confidence level to use for confidence intervals. |
times |
A vector of times to use for survival summaries. |
levels |
A vector of levels that character |
by |
a vector of the by-values. |
by.levels |
a vector of the levels of |
totallab |
What to call the total "column" |
Not all these functions are exported, in order to avoid conflicting NAMESPACES.
Note also that the functions prefixed with "arsenal_" can be referred to by their short names
(e.g., "min" for "arsenal_min").
Usually a vector of the appropriate numbers.
tableby functionsA collection of functions that may help users create custom functions that are formatted correctly.
tbfmt(x, digits = NULL, digits.count = NULL, digits.pct = NULL, ...) as.tbstat( x, oldClass = NULL, fmt = "{y}", which.count = 0L, which.pct = 0L, ... ) as.tbstat_multirow(x)tbfmt(x, digits = NULL, digits.count = NULL, digits.pct = NULL, ...) as.tbstat( x, oldClass = NULL, fmt = "{y}", which.count = 0L, which.pct = 0L, ... ) as.tbstat_multirow(x)
x |
Usually a vector. |
digits, digits.pct, digits.count
|
Digits specifications |
... |
arguments to pass to |
oldClass |
class(es) to add to the resulting object. |
fmt |
A |
which.count |
Which statistics are counts? The default is 0, indicating that none are. |
which.pct |
Which statistics are percents? The default is 0, indicating that none are. |
The vignette has an example on how to use these.
as.tbstat defines a tableby statistic with its appropriate formatting.
tbfmt applies some default formatting.
as.tbstat_multirow marks an object (usually a list) for multiple-row printing.
Functions to output tables to a single document. (Also the S3 backbone behind the write2* functions.)
write2(object, file, ..., output_format) ## S3 method for class 'arsenal_table' write2(object, file, ..., output_format = NULL) ## S3 method for class 'summary.arsenal_table' write2(object, file, ..., output_format = NULL) ## S3 method for class 'comparedf' write2(object, file, ..., output_format = NULL) ## S3 method for class 'summary.comparedf' write2(object, file, ..., output_format = NULL) ## S3 method for class 'verbatim' write2(object, file, ..., output_format = NULL) ## S3 method for class 'yaml' write2(object, file, ..., output_format = NULL) ## S3 method for class 'code.chunk' write2(object, file, ..., output_format = NULL) ## S3 method for class 'knitr_kable' write2(object, file, ..., output_format = NULL) ## S3 method for class 'xtable' write2(object, file, ..., output_format = NULL) ## S3 method for class 'character' write2(object, file, ..., output_format = NULL) ## S3 method for class 'list' write2( object, file, ..., append. = FALSE, render. = TRUE, keep.rmd = !render., output_format = NULL ) ## Default S3 method: write2( object, file, FUN = NULL, ..., append. = FALSE, render. = TRUE, keep.rmd = !render., output_format = NULL )write2(object, file, ..., output_format) ## S3 method for class 'arsenal_table' write2(object, file, ..., output_format = NULL) ## S3 method for class 'summary.arsenal_table' write2(object, file, ..., output_format = NULL) ## S3 method for class 'comparedf' write2(object, file, ..., output_format = NULL) ## S3 method for class 'summary.comparedf' write2(object, file, ..., output_format = NULL) ## S3 method for class 'verbatim' write2(object, file, ..., output_format = NULL) ## S3 method for class 'yaml' write2(object, file, ..., output_format = NULL) ## S3 method for class 'code.chunk' write2(object, file, ..., output_format = NULL) ## S3 method for class 'knitr_kable' write2(object, file, ..., output_format = NULL) ## S3 method for class 'xtable' write2(object, file, ..., output_format = NULL) ## S3 method for class 'character' write2(object, file, ..., output_format = NULL) ## S3 method for class 'list' write2( object, file, ..., append. = FALSE, render. = TRUE, keep.rmd = !render., output_format = NULL ) ## Default S3 method: write2( object, file, FUN = NULL, ..., append. = FALSE, render. = TRUE, keep.rmd = !render., output_format = NULL )
object |
An object. |
file |
A single character string denoting the filename for the output document. |
... |
Additional arguments to be passed to |
output_format |
One of the following:
See |
append. |
Logical, denoting whether (if a temporary |
render. |
Logical, denoting whether to render the temporary |
keep.rmd |
Logical, denoting whether to keep the intermediate |
FUN |
The summary-like or print-like function to use to generate the markdown content. Can be passed as a function or a
character string. It's expected that |
write2 is an S3 method. The default prints the object (using print)
inside a section surrounded by three back ticks. See verbatim for details.
There are methods implemented for tableby, modelsum, and freqlist, all of which use the
summary function. There are also methods compatible with kable, xtable,
and pander_return. Another option is to coerce an object using verbatim() to print out the
results monospaced (as if they were in the terminal). To output multiple tables into a document, simply make a list of them
and call the same function as before. Finally, to output code chunks to be evaluated, use code.chunk.
For more information, see vignette("write2").
object is returned invisibly, and file is written.
Ethan Heinzen, adapted from code from Krista Goergen
write2word, write2pdf, write2html,
render, word_document, html_document, pdf_document,
rtf_document, md_document, odt_document
## Not run: data(mockstudy) # tableby example tab1 <- tableby(arm ~ sex + age, data=mockstudy) write2(tab1, tempfile(fileext = ".rtf"), toc = TRUE, # passed to rmarkdown::rtf_document, though in this case it's not practical quiet = TRUE, # passed to rmarkdown::render title = "My cool new title", # passed to summary.tableby output_format = rmarkdown::rtf_document) write2html(list( "# Header 1", # a header code.chunk(a <- 1, b <- 2, a + b), # a code chunk verbatim("hi there") # verbatim output ), tempfile(fileext = ".html"), quite = TRUE) ## End(Not run)## Not run: data(mockstudy) # tableby example tab1 <- tableby(arm ~ sex + age, data=mockstudy) write2(tab1, tempfile(fileext = ".rtf"), toc = TRUE, # passed to rmarkdown::rtf_document, though in this case it's not practical quiet = TRUE, # passed to rmarkdown::render title = "My cool new title", # passed to summary.tableby output_format = rmarkdown::rtf_document) write2html(list( "# Header 1", # a header code.chunk(a <- 1, b <- 2, a + b), # a code chunk verbatim("hi there") # verbatim output ), tempfile(fileext = ".html"), quite = TRUE) ## End(Not run)
write2
Helper functions for write2.
verbatim(...) code.chunk(..., chunk.opts = "r")verbatim(...) code.chunk(..., chunk.opts = "r")
... |
For |
chunk.opts |
A single character string giving the code chunk options. Make sure to specify the engine! |
The "verbatim" class is to tell write2 to print the object inside
a section surrounded by three back ticks. The results will look like it would in the terminal (monospaced).
code.chunk() is to write explicit code chunks in the .Rmd file; it captures the call and writes it to the
file, to execute upon knitting.
Functions to output tables to a single Word, HTML, or PDF document.
write2word(object, file, ...) write2pdf(object, file, ...) write2html(object, file, ...)write2word(object, file, ...) write2pdf(object, file, ...) write2html(object, file, ...)
object |
An object. |
file |
A single character string denoting the filename for the output document. |
... |
Additional arguments to be passed to |
To generate the appropriate file type, the write2* functions use one of rmarkdown::word_document, rmarkdown::html_document,
and rmarkdown::pdf_document to get the job done. "..." arguments are passed to these functions, too.
object is returned invisibly, and file is written.
Ethan Heinzen, adapted from code from Krista Goergen
## Not run: data(mockstudy) # tableby example tab1 <- tableby(arm ~ sex + age, data=mockstudy) write2html(tab1, "~/trash.html") # freqlist example tab.ex <- table(mockstudy[, c("arm", "sex", "mdquality.s")], useNA = "ifany") noby <- freqlist(tab.ex, na.options = "include") write2pdf(noby, "~/trash2.pdf") # A more complicated example write2word(tab1, "~/trash.doc", keep.md = TRUE, reference_docx = mystyles.docx, # passed to rmarkdown::word_document quiet = TRUE, # passed to rmarkdown::render title = "My cool new title") # passed to summary.tableby ## End(Not run)## Not run: data(mockstudy) # tableby example tab1 <- tableby(arm ~ sex + age, data=mockstudy) write2html(tab1, "~/trash.html") # freqlist example tab.ex <- table(mockstudy[, c("arm", "sex", "mdquality.s")], useNA = "ifany") noby <- freqlist(tab.ex, na.options = "include") write2pdf(noby, "~/trash2.pdf") # A more complicated example write2word(tab1, "~/trash.doc", keep.md = TRUE, reference_docx = mystyles.docx, # passed to rmarkdown::word_document quiet = TRUE, # passed to rmarkdown::render title = "My cool new title") # passed to summary.tableby ## End(Not run)
write2
Include a YAML header in write2
yaml(...) ## S3 method for class 'yaml' print(x, ...) ## S3 method for class 'yaml' c(..., recursive = FALSE) is.yaml(x)yaml(...) ## S3 method for class 'yaml' print(x, ...) ## S3 method for class 'yaml' c(..., recursive = FALSE) is.yaml(x)
... |
For |
x |
An object of class |
recursive |
Not in use at this time. |
A text string of class "yaml".
Ethan Heinzen, adapted from an idea by Brendan Broderick
x <- yaml(title = "My cool title", author = "Ethan P Heinzen") x y <- yaml("header-includes" = list("\\usepackage[labelformat=empty]{caption}")) y c(x, y)x <- yaml(title = "My cool title", author = "Ethan P Heinzen") x y <- yaml("header-includes" = list("\\usepackage[labelformat=empty]{caption}")) y c(x, y)