732 lines
19 KiB
Plaintext
732 lines
19 KiB
Plaintext
|
%\VignetteIndexEntry{xtable Gallery}
|
||
|
|
%\VignetteDepends{xtable}
|
||
|
|
%\VignetteKeywords{LaTeX, HTML, table}
|
||
|
|
%\VignettePackage{xtable}
|
||
|
|
% !Rnw weave = knitr
|
||
|
|
% \VignetteEngine{knitr::knitr}
|
||
|
|
%**************************************************************************
|
||
|
|
\documentclass{article}
|
||
|
|
\usepackage[a4paper,height=24cm]{geometry} % geometry first
|
||
|
|
\usepackage{array}
|
||
|
|
\usepackage{booktabs}
|
||
|
|
\usepackage{longtable}
|
||
|
|
\usepackage{parskip}
|
||
|
|
\usepackage{rotating}
|
||
|
|
\usepackage{tabularx}
|
||
|
|
\usepackage{titlesec}
|
||
|
|
\usepackage{hyperref} % hyperref last
|
||
|
|
\titleformat\subsubsection{\bfseries\itshape}{}{0pt}{}
|
||
|
|
\newcommand\p{\vspace{2ex}}
|
||
|
|
\newcommand\code[1]{\texttt{#1}}
|
||
|
|
\newcommand\pkg[1]{\textbf{#1}}
|
||
|
|
\setcounter{tocdepth}{2}
|
||
|
|
\begin{document}
|
||
|
|
|
||
|
|
\title{The \pkg{xtable} Gallery}
|
||
|
|
\author{Jonathan Swinton and others}
|
||
|
|
\maketitle
|
||
|
|
|
||
|
|
\tableofcontents
|
||
|
|
|
||
|
|
\newpage
|
||
|
|
|
||
|
|
\section{Introduction}
|
||
|
|
This document gives a gallery of tables which can be made using the
|
||
|
|
\pkg{xtable} package to create \LaTeX\ output. It doubles as a
|
||
|
|
regression check for the package.
|
||
|
|
|
||
|
|
<<include=FALSE>>=
|
||
|
|
library(knitr)
|
||
|
|
opts_chunk$set(fig.path='figdir/fig', debug=TRUE, echo=TRUE)
|
||
|
|
set.seed(1234)
|
||
|
|
@
|
||
|
|
|
||
|
|
The first step is to load the package and set an option for this document.
|
||
|
|
<<results='asis'>>=
|
||
|
|
library(xtable)
|
||
|
|
options(xtable.floating = FALSE)
|
||
|
|
options(xtable.timestamp = "")
|
||
|
|
@
|
||
|
|
|
||
|
|
\section{Gallery}
|
||
|
|
\subsection{Data frame}
|
||
|
|
<<results='asis'>>=
|
||
|
|
data(tli)
|
||
|
|
xtable(tli[1:10, ])
|
||
|
|
@
|
||
|
|
|
||
|
|
\subsection{Matrix}
|
||
|
|
<<results='asis'>>=
|
||
|
|
design.matrix <- model.matrix(~ sex*grade, data = tli[1:10, ])
|
||
|
|
xtable(design.matrix, digits = 0)
|
||
|
|
@
|
||
|
|
|
||
|
|
\newpage
|
||
|
|
\subsection{aov}
|
||
|
|
<<results='asis'>>=
|
||
|
|
fm1 <- aov(tlimth ~ sex + ethnicty + grade + disadvg, data = tli)
|
||
|
|
xtable(fm1)
|
||
|
|
@
|
||
|
|
|
||
|
|
\subsection{lm}
|
||
|
|
<<results='asis'>>=
|
||
|
|
fm2 <- lm(tlimth ~ sex*ethnicty, data = tli)
|
||
|
|
xtable(fm2)
|
||
|
|
@
|
||
|
|
|
||
|
|
\subsubsection{Anova table (one model)}
|
||
|
|
<<results='asis'>>=
|
||
|
|
xtable(anova(fm2))
|
||
|
|
@
|
||
|
|
|
||
|
|
\subsubsection{Anova table (two models)}
|
||
|
|
<<results='asis'>>=
|
||
|
|
fm2b <- lm(tlimth ~ ethnicty, data = tli)
|
||
|
|
xtable(anova(fm2b, fm2))
|
||
|
|
@
|
||
|
|
|
||
|
|
\subsubsection{Anova list}
|
||
|
|
|
||
|
|
<<aovlist>>=
|
||
|
|
Block <- gl(8, 4)
|
||
|
|
A <- factor(c(0,1,0,1,0,1,0,1,0,1,0,1,0,1,0,1,0,1,0,1,
|
||
|
|
0,1,0,1,0,1,0,1,0,1,0,1))
|
||
|
|
B <- factor(c(0,0,1,1,0,0,1,1,0,1,0,1,1,0,1,0,0,0,1,1,
|
||
|
|
0,0,1,1,0,0,1,1,0,0,1,1))
|
||
|
|
C <- factor(c(0,1,1,0,1,0,0,1,0,0,1,1,0,0,1,1,0,1,0,1,
|
||
|
|
1,0,1,0,0,0,1,1,1,1,0,0))
|
||
|
|
Yield <- c(101, 373, 398, 291, 312, 106, 265, 450, 106, 306, 324, 449,
|
||
|
|
272, 89, 407, 338, 87, 324, 279, 471, 323, 128, 423, 334,
|
||
|
|
131, 103, 445, 437, 324, 361, 302, 272)
|
||
|
|
aovdat <- data.frame(Block, A, B, C, Yield)
|
||
|
|
|
||
|
|
old <- getOption("contrasts")
|
||
|
|
options(contrasts = c("contr.helmert", "contr.poly"))
|
||
|
|
(fit <- aov(Yield ~ A*B*C + Error(Block), data = aovdat))
|
||
|
|
class(fit)
|
||
|
|
summary(fit)
|
||
|
|
options(contrasts = old)
|
||
|
|
@
|
||
|
|
|
||
|
|
\p
|
||
|
|
|
||
|
|
<<xtableaovlist, results='asis'>>=
|
||
|
|
xtable(fit)
|
||
|
|
@
|
||
|
|
|
||
|
|
|
||
|
|
\newpage
|
||
|
|
\subsection{glm}
|
||
|
|
<<results='asis'>>=
|
||
|
|
fm3 <- glm(disadvg ~ ethnicty*grade, data = tli, family = binomial)
|
||
|
|
xtable(fm3)
|
||
|
|
@
|
||
|
|
|
||
|
|
\subsubsection{Analysis of deviance}
|
||
|
|
<<results='asis'>>=
|
||
|
|
xtable(anova(fm3))
|
||
|
|
@
|
||
|
|
|
||
|
|
\subsection{prcomp}
|
||
|
|
<<results='asis'>>=
|
||
|
|
pr1 <- prcomp(USArrests)
|
||
|
|
xtable(pr1)
|
||
|
|
@
|
||
|
|
|
||
|
|
\p
|
||
|
|
<<results='asis'>>=
|
||
|
|
xtable(summary(pr1))
|
||
|
|
@
|
||
|
|
|
||
|
|
<<include=FALSE>>=
|
||
|
|
# pr2 <- princomp(USArrests)
|
||
|
|
# xtable(pr2)
|
||
|
|
@
|
||
|
|
|
||
|
|
\newpage
|
||
|
|
|
||
|
|
\subsection{Time series}
|
||
|
|
<<results='asis'>>=
|
||
|
|
temp.ts <- ts(cumsum(1 + round(rnorm(100), 0)),
|
||
|
|
start = c(1954, 7), frequency = 12)
|
||
|
|
temp.table <- xtable(temp.ts, digits = 0)
|
||
|
|
temp.table
|
||
|
|
@
|
||
|
|
|
||
|
|
\subsection{Flat tables}
|
||
|
|
\label{sec:flat-tables}
|
||
|
|
|
||
|
|
See the \textbf{Details} section of the help for \code{ftable} for a
|
||
|
|
description of these tables, which are flat versions of
|
||
|
|
multi-dimensional contingency tables. They require special methods to
|
||
|
|
enable them to be printed using \pkg{xtable}.
|
||
|
|
|
||
|
|
|
||
|
|
<<ftable>>=
|
||
|
|
data(mtcars)
|
||
|
|
mtcars$cyl <- factor(mtcars$cyl, levels = c("4","6","8"),
|
||
|
|
labels = c("four","six","eight"))
|
||
|
|
tbl <- ftable(mtcars$cyl, mtcars$vs, mtcars$am, mtcars$gear,
|
||
|
|
row.vars = c(2, 4),
|
||
|
|
dnn = c("Cylinders", "V/S", "Transmission", "Gears"))
|
||
|
|
tbl
|
||
|
|
@ %def
|
||
|
|
|
||
|
|
Here is the \LaTeX{} produced:
|
||
|
|
|
||
|
|
<<ftablecheck>>=
|
||
|
|
xftbl <- xtableFtable(tbl, method = "compact")
|
||
|
|
print.xtableFtable(xftbl, booktabs = TRUE)
|
||
|
|
@ %def
|
||
|
|
|
||
|
|
And here is a basic flat table:
|
||
|
|
|
||
|
|
<<ftable1, results = 'asis'>>=
|
||
|
|
xftbl <- xtableFtable(tbl)
|
||
|
|
print.xtableFtable(xftbl)
|
||
|
|
@ %def
|
||
|
|
|
||
|
|
This illustrates the \code{method} argument:
|
||
|
|
|
||
|
|
<<ftable2, results = 'asis'>>=
|
||
|
|
xftbl <- xtableFtable(tbl, method = "col.compact")
|
||
|
|
print.xtableFtable(xftbl, rotate.rownames = TRUE)
|
||
|
|
@ %def
|
||
|
|
|
||
|
|
Booktabs is incompatible with vertical lines in tables, so the
|
||
|
|
vertical dividing line is removed.
|
||
|
|
|
||
|
|
<<ftable3, results = 'asis'>>=
|
||
|
|
xftbl <- xtableFtable(tbl, method = "compact")
|
||
|
|
print.xtableFtable(xftbl, booktabs = TRUE)
|
||
|
|
@ %def
|
||
|
|
\p
|
||
|
|
|
||
|
|
Row and column variable names can be formatted specially using
|
||
|
|
sanitization, and row and column variable names and labels can be
|
||
|
|
rotated.
|
||
|
|
|
||
|
|
If special formatting is required for row and column labels, that can
|
||
|
|
be done as a workaround by redefining the data and associated labels.
|
||
|
|
|
||
|
|
<<ftable4, results = 'asis'>>=
|
||
|
|
italic <- function(x){
|
||
|
|
paste0('{\\emph{', x, '}}')
|
||
|
|
}
|
||
|
|
mtcars$cyl <- factor(mtcars$cyl, levels = c("four","six","eight"),
|
||
|
|
labels = c("four",italic("six"),"eight"))
|
||
|
|
large <- function(x){
|
||
|
|
paste0('{\\Large ', x, '}')
|
||
|
|
}
|
||
|
|
bold <- function(x){
|
||
|
|
paste0('{\\bfseries ', x, '}')
|
||
|
|
}
|
||
|
|
tbl <- ftable(mtcars$cyl, mtcars$vs, mtcars$am, mtcars$gear,
|
||
|
|
row.vars = c(2, 4),
|
||
|
|
dnn = c("Cylinders", "V/S", "Transmission", "Gears"))
|
||
|
|
xftbl <- xtableFtable(tbl, method = "row.compact")
|
||
|
|
print.xtableFtable(xftbl,
|
||
|
|
sanitize.rownames.function = large,
|
||
|
|
sanitize.colnames.function = bold,
|
||
|
|
rotate.colnames = TRUE,
|
||
|
|
rotate.rownames = TRUE)
|
||
|
|
@ %def
|
||
|
|
|
||
|
|
|
||
|
|
|
||
|
|
\newpage
|
||
|
|
|
||
|
|
<<include=FALSE>>=
|
||
|
|
# ## Demonstrate saving to file
|
||
|
|
# for(i in c("latex", "html")) {
|
||
|
|
# outFileName <- paste("xtable.", ifelse(i=="latex", "tex", i), sep = "")
|
||
|
|
# print(xtable(lm.D9), type = i, file = outFileName, append = TRUE,
|
||
|
|
# latex.environments = NULL)
|
||
|
|
# print(xtable(lm.D9), type = i, file = outFileName, append = TRUE,
|
||
|
|
# latex.environments = "")
|
||
|
|
# print(xtable(lm.D9), type = i, file = outFileName, append = TRUE,
|
||
|
|
# latex.environments = "center")
|
||
|
|
# print(xtable(anova(glm.D93, test = "Chisq")),
|
||
|
|
# type = i, file = outFileName,
|
||
|
|
# append = TRUE)
|
||
|
|
# print(xtable(anova(glm.D93)), hline.after = c(1),
|
||
|
|
# size = "small", type = i,
|
||
|
|
# file = outFileName, append = TRUE)
|
||
|
|
# # print(xtable(pr2), type = i, file = outFileName, append = TRUE)
|
||
|
|
# }
|
||
|
|
@
|
||
|
|
|
||
|
|
\section{Automatic formatting}
|
||
|
|
\subsection{Suggest alignment, digits, and display}
|
||
|
|
The functions \code{xalign}, \code{xdigits}, and \code{xdisplay} are
|
||
|
|
useful for formatting tables in a sensible way. Consider the output
|
||
|
|
produced by the default formatting.
|
||
|
|
|
||
|
|
<<results='asis'>>=
|
||
|
|
data(mtcars)
|
||
|
|
dat <- mtcars[1:3, 1:6]
|
||
|
|
x <- xtable(dat)
|
||
|
|
x
|
||
|
|
@
|
||
|
|
|
||
|
|
\p
|
||
|
|
Now change the default alignment, digits and display using helper functions
|
||
|
|
\code{xalign}, \code{xdigits}, and \code{xdisplay}. This produces a better
|
||
|
|
format as shown below.
|
||
|
|
|
||
|
|
<<results='asis'>>=
|
||
|
|
align(x) <- xalign(x)
|
||
|
|
digits(x) <- xdigits(x)
|
||
|
|
display(x) <- xdisplay(x)
|
||
|
|
x
|
||
|
|
@
|
||
|
|
|
||
|
|
\subsection{Shorthand notation}
|
||
|
|
For convenience, the three `autoformat' functions (\code{xalign},
|
||
|
|
\code{xdigits}, and \code{xdisplay}) can be applied together when an
|
||
|
|
\code{xtable} is created, using the \code{auto} argument:
|
||
|
|
|
||
|
|
<<results='asis'>>=
|
||
|
|
xtable(dat, auto = TRUE)
|
||
|
|
@
|
||
|
|
|
||
|
|
\p
|
||
|
|
Similarly, the \code{autoformat} function can be used to postprocess an
|
||
|
|
existing \code{xtable}:
|
||
|
|
|
||
|
|
<<results='asis'>>=
|
||
|
|
x <- xtable(dat)
|
||
|
|
autoformat(x)
|
||
|
|
@
|
||
|
|
|
||
|
|
\newpage
|
||
|
|
|
||
|
|
\subsection{Math-Style Exponents}
|
||
|
|
If you prefer $5 \times 10^5$ in your tables to 5e5, the
|
||
|
|
\code{math.style.exponents} option to \code{print.xtable} is useful:
|
||
|
|
|
||
|
|
<<results='asis'>>=
|
||
|
|
print(xtable(data.frame(text = c("foo","bar"),
|
||
|
|
googols = c(10e10,50e10),
|
||
|
|
small = c(8e-24,7e-5),
|
||
|
|
row.names = c("A","B")),
|
||
|
|
display = c("s","s","g","g")),
|
||
|
|
math.style.exponents = TRUE)
|
||
|
|
@
|
||
|
|
|
||
|
|
This option also supports the values \code{ensuremath} which uses
|
||
|
|
\code{\char`\\ensuremath} instead of \code{\$\$} and \code{UTF-8}
|
||
|
|
which uses UTF-8 to approximate the \LaTeX\ typesetting.
|
||
|
|
|
||
|
|
|
||
|
|
\section{Sanitization}
|
||
|
|
<<results='asis'>>=
|
||
|
|
insane <- data.frame(Name = c("Ampersand","Greater than","Less than",
|
||
|
|
"Underscore","Per cent","Dollar",
|
||
|
|
"Backslash","Hash","Caret","Tilde",
|
||
|
|
"Left brace","Right brace"),
|
||
|
|
Character = I(c("&",">","<","_","%","$",
|
||
|
|
"\\","#","^","~","{","}")))
|
||
|
|
colnames(insane)[2] <- paste(insane[, 2], collapse = "")
|
||
|
|
xtable(insane)
|
||
|
|
@
|
||
|
|
|
||
|
|
\p
|
||
|
|
Sometimes you might want to have your own sanitization function.
|
||
|
|
|
||
|
|
|
||
|
|
<<results='asis'>>=
|
||
|
|
wanttex <- xtable(data.frame(Column =
|
||
|
|
paste("Value_is $10^{-",1:3,"}$", sep = "")))
|
||
|
|
print(wanttex, sanitize.text.function =
|
||
|
|
function(str) gsub("_", "\\_", str, fixed = TRUE))
|
||
|
|
@
|
||
|
|
|
||
|
|
\p
|
||
|
|
Sanitization can be useful in formatting column headings and row names:
|
||
|
|
|
||
|
|
<<sanitize3>>=
|
||
|
|
dat <- mtcars[1:3, 1:6]
|
||
|
|
large <- function(x){
|
||
|
|
paste0('{\\Large{\\bfseries ', x, '}}')
|
||
|
|
}
|
||
|
|
italic <- function(x){
|
||
|
|
paste0('{\\emph{ ', x, '}}')
|
||
|
|
}
|
||
|
|
@ %def
|
||
|
|
|
||
|
|
<<sanitize4, results = 'asis'>>=
|
||
|
|
print(xtable(dat),
|
||
|
|
sanitize.rownames.function = italic,
|
||
|
|
sanitize.colnames.function = large,
|
||
|
|
booktabs = TRUE)
|
||
|
|
@ %def
|
||
|
|
|
||
|
|
|
||
|
|
|
||
|
|
\newpage
|
||
|
|
|
||
|
|
\subsection{Markup in tables}
|
||
|
|
Markup can be included in tables, including in column and row names,
|
||
|
|
by using a custom \code{sanitize.text.function}.
|
||
|
|
|
||
|
|
<<results='asis'>>=
|
||
|
|
mat <- round(matrix(c(0.9, 0.89, 200, 0.045, 2.0), c(1, 5)), 4)
|
||
|
|
rownames(mat) <- "$y_{t-1}$"
|
||
|
|
colnames(mat) <- c("$R^2$", "$\\bar{x}$", "F-stat", "S.E.E", "DW")
|
||
|
|
mat <- xtable(mat)
|
||
|
|
print(mat, sanitize.text.function = function(x) {x})
|
||
|
|
@
|
||
|
|
|
||
|
|
% By David Dahl to demonstrate contribution from David Whitting, 2007-10-09.
|
||
|
|
\p
|
||
|
|
You can also have sanitize functions that are specific to column or
|
||
|
|
row names. In the table below, the row name is not sanitized but
|
||
|
|
column names and table elements are.
|
||
|
|
|
||
|
|
<<results='asis'>>=
|
||
|
|
money <- matrix(c("$1,000","$900","$100"), ncol = 3,
|
||
|
|
dimnames = list("$\\alpha$",
|
||
|
|
c("Income (US$)","Expenses (US$)",
|
||
|
|
"Profit (US$)")))
|
||
|
|
print(xtable(money), sanitize.rownames.function = function(x) {x})
|
||
|
|
@
|
||
|
|
|
||
|
|
\section{Format examples}
|
||
|
|
\subsection{Adding a centering environment}
|
||
|
|
<<results='asis'>>=
|
||
|
|
print(xtable(anova(fm3), caption = "\\tt latex.environments = \"\""),
|
||
|
|
floating = TRUE, latex.environments = "")
|
||
|
|
print(xtable(anova(fm3), caption = "\\tt latex.environments = \"center\""),
|
||
|
|
floating = TRUE, latex.environments = "center")
|
||
|
|
@
|
||
|
|
|
||
|
|
\newpage
|
||
|
|
|
||
|
|
\subsection{Column alignment}
|
||
|
|
<<results='asis'>>=
|
||
|
|
tli.table <- xtable(tli[1:10, ])
|
||
|
|
align(tli.table) <- rep("r", 6)
|
||
|
|
tli.table
|
||
|
|
@
|
||
|
|
|
||
|
|
\subsubsection{Left aligned strings with column lines}
|
||
|
|
<<results='asis'>>=
|
||
|
|
align(tli.table) <- "|rrl|l|lr|"
|
||
|
|
tli.table
|
||
|
|
@
|
||
|
|
|
||
|
|
\subsubsection{Fixed width columns}
|
||
|
|
<<results='asis'>>=
|
||
|
|
align(tli.table) <- "|rr|lp{3cm}l|r|"
|
||
|
|
tli.table
|
||
|
|
@
|
||
|
|
|
||
|
|
\newpage
|
||
|
|
|
||
|
|
\subsection{Number of digits}
|
||
|
|
One number for all columns,
|
||
|
|
<<results='asis'>>=
|
||
|
|
display(tli.table)[c(2,6)] <- "f"
|
||
|
|
digits(tli.table) <- 3
|
||
|
|
tli.table
|
||
|
|
@
|
||
|
|
|
||
|
|
\p
|
||
|
|
or one for each column, including the row names,
|
||
|
|
<<results='asis'>>=
|
||
|
|
digits(tli.table) <- 1:(ncol(tli)+1)
|
||
|
|
tli.table
|
||
|
|
@
|
||
|
|
|
||
|
|
\p
|
||
|
|
or as a full matrix.
|
||
|
|
<<results='asis'>>=
|
||
|
|
digits(tli.table) <- matrix(0:4, nrow = 10, ncol = ncol(tli)+1)
|
||
|
|
tli.table
|
||
|
|
@
|
||
|
|
|
||
|
|
\newpage
|
||
|
|
|
||
|
|
\subsection{Suppress row/column names}
|
||
|
|
\subsubsection{Suppress row names}
|
||
|
|
<<results='asis'>>=
|
||
|
|
tli.table <- xtable(tli[1:10, ])
|
||
|
|
print(tli.table, include.rownames = FALSE)
|
||
|
|
@
|
||
|
|
|
||
|
|
\p
|
||
|
|
If you want a vertical line on the left, you need to change the \code{align}
|
||
|
|
attribute.
|
||
|
|
<<results='asis'>>=
|
||
|
|
align(tli.table) <- "|r|r|lp{3cm}l|r|"
|
||
|
|
print(tli.table, include.rownames = FALSE)
|
||
|
|
@
|
||
|
|
|
||
|
|
\p
|
||
|
|
Revert the alignment to what is was before.
|
||
|
|
<<>>=
|
||
|
|
align(tli.table) <- "|rr|lp{3cm}l|r|"
|
||
|
|
@
|
||
|
|
|
||
|
|
\newpage
|
||
|
|
|
||
|
|
\subsubsection{Suppress column names}
|
||
|
|
<<results='asis'>>=
|
||
|
|
print(tli.table, include.colnames = FALSE)
|
||
|
|
@
|
||
|
|
|
||
|
|
\p
|
||
|
|
Note the doubled header lines which can be suppressed.
|
||
|
|
<<results='asis'>>=
|
||
|
|
print(tli.table, include.colnames = FALSE,
|
||
|
|
hline.after = c(0,nrow(tli.table)))
|
||
|
|
@
|
||
|
|
|
||
|
|
\subsubsection{Suppress row and column names}
|
||
|
|
<<results='asis'>>=
|
||
|
|
print(tli.table, include.colnames = FALSE, include.rownames = FALSE)
|
||
|
|
@
|
||
|
|
|
||
|
|
\newpage
|
||
|
|
|
||
|
|
\subsection{Rotate row/column names}
|
||
|
|
The \code{rotate.rownames} and \code{rotate.colnames} arguments can be
|
||
|
|
used to rotate the row and/or column names. This requires
|
||
|
|
\verb|\usepackage{rotating}| in the \LaTeX\ preamble.
|
||
|
|
|
||
|
|
<<results='asis'>>=
|
||
|
|
print(tli.table, rotate.rownames = TRUE, rotate.colnames = TRUE)
|
||
|
|
@
|
||
|
|
|
||
|
|
\newpage
|
||
|
|
|
||
|
|
\subsection{Horizontal lines}
|
||
|
|
\subsubsection{Line locations}
|
||
|
|
Use the \code{hline.after} argument to specify the position of the
|
||
|
|
horizontal lines.
|
||
|
|
|
||
|
|
<<results='asis'>>=
|
||
|
|
print(xtable(anova(fm3)), hline.after = c(1))
|
||
|
|
@
|
||
|
|
|
||
|
|
\subsubsection{Line styles}
|
||
|
|
Specifying \code{booktabs = TRUE} will generate three line types. By
|
||
|
|
default, when no value is given for \code{hline.after}, a
|
||
|
|
\verb|\toprule| will be drawn above the table, a \verb|\midrule| after
|
||
|
|
the table headings and a \verb|\bottomrule| below the table. This
|
||
|
|
requires \verb|\usepackage{booktabs}| in the \LaTeX\ preamble.
|
||
|
|
|
||
|
|
\p
|
||
|
|
|
||
|
|
The top and bottom rules are slightly thicker than the mid rule. The
|
||
|
|
thickness of the lines can be set via the \LaTeX\ lengths
|
||
|
|
\verb|\heavyrulewidth| and \verb|\lightrulewidth|.
|
||
|
|
|
||
|
|
<<results='asis'>>=
|
||
|
|
tli.table <- xtable(tli[1:10, ])
|
||
|
|
print(tli.table, include.rownames = FALSE, booktabs = TRUE)
|
||
|
|
@
|
||
|
|
|
||
|
|
\p
|
||
|
|
|
||
|
|
If \code{hline.after} includes \code{-1}, a \verb|\toprule| will be
|
||
|
|
drawn above the table. If \code{hline.after} includes the number of
|
||
|
|
rows in the table, a \verb|\bottomrule| will be drawn below the
|
||
|
|
table. For any other values specified in \code{hline.after}, a
|
||
|
|
\verb|\midrule| will be drawn after that line of the table.
|
||
|
|
|
||
|
|
\p
|
||
|
|
The following table has more than one \verb|\midrule|.
|
||
|
|
|
||
|
|
<<results='asis'>>=
|
||
|
|
bktbs <- xtable(matrix(1:10, ncol = 2))
|
||
|
|
hlines <- c(-1, 0, 1, nrow(bktbs))
|
||
|
|
print(bktbs, booktabs = TRUE, hline.after = hlines)
|
||
|
|
@
|
||
|
|
|
||
|
|
\subsection{Table level commands}
|
||
|
|
<<results='asis'>>=
|
||
|
|
print(xtable(anova(fm3)), size = "large")
|
||
|
|
@
|
||
|
|
|
||
|
|
\p
|
||
|
|
<<results='asis'>>=
|
||
|
|
print(xtable(anova(fm3)), size = "\\setlength{\\tabcolsep}{12pt}")
|
||
|
|
@
|
||
|
|
|
||
|
|
\subsection{Long tables}
|
||
|
|
Requires \verb|\usepackage{longtable}| in the \LaTeX\ preamble.
|
||
|
|
|
||
|
|
<<results='asis'>>=
|
||
|
|
x <- matrix(rnorm(1000), ncol = 10)
|
||
|
|
x.big <- xtable(x, caption = "A \\code{longtable} spanning several pages")
|
||
|
|
print(x.big, hline.after=c(-1, 0), tabular.environment = "longtable")
|
||
|
|
@
|
||
|
|
|
||
|
|
Extra features of the \pkg{longtable} \LaTeX{} package can typically
|
||
|
|
be activated using \code{add.to.row}, as shown below.
|
||
|
|
|
||
|
|
<<results='asis'>>=
|
||
|
|
add.to.row <- list(pos = list(0), command = NULL)
|
||
|
|
command <- paste0("\\hline\n\\endhead\n",
|
||
|
|
"\\hline\n",
|
||
|
|
"\\multicolumn{", dim(x)[2] + 1, "}{l}",
|
||
|
|
"{\\footnotesize Continued on next page}\n",
|
||
|
|
"\\endfoot\n",
|
||
|
|
"\\endlastfoot\n")
|
||
|
|
add.to.row$command <- command
|
||
|
|
print(x.big, hline.after=c(-1), add.to.row = add.to.row,
|
||
|
|
tabular.environment = "longtable")
|
||
|
|
@
|
||
|
|
|
||
|
|
|
||
|
|
\newpage
|
||
|
|
|
||
|
|
\subsection{Use of \code{add.to.row} argument}
|
||
|
|
The following frequency table has outer dimnames: \code{Grade3} and
|
||
|
|
\code{Grade6}.
|
||
|
|
|
||
|
|
<<>>=
|
||
|
|
Grade3 <- c("A","B","B","A","B","C","C","D","A","B",
|
||
|
|
"C","C","C","D","B","B","D","C","C","D")
|
||
|
|
Grade6 <- c("A","A","A","B","B","B","B","B","C","C",
|
||
|
|
"A","C","C","C","D","D","D","D","D","D")
|
||
|
|
Cohort <- table(Grade3, Grade6)
|
||
|
|
Cohort
|
||
|
|
@
|
||
|
|
|
||
|
|
\p
|
||
|
|
The default behavior of \code{print.xtable} is to strip outer dimnames.
|
||
|
|
<<results='asis'>>=
|
||
|
|
xtable(Cohort)
|
||
|
|
@
|
||
|
|
|
||
|
|
\p
|
||
|
|
The desired column labels can be created using \code{add.to.row}, in this case
|
||
|
|
applying two commands to ``row number zero'' while suppressing the basic column
|
||
|
|
names.
|
||
|
|
|
||
|
|
<<results='asis'>>=
|
||
|
|
addtorow <- list()
|
||
|
|
addtorow$pos <- list(0, 0)
|
||
|
|
addtorow$command <- c("& \\multicolumn{4}{c}{Grade 6} \\\\\n",
|
||
|
|
"Grade 3 & A & B & C & D \\\\\n")
|
||
|
|
print(xtable(Cohort), add.to.row = addtorow, include.colnames = FALSE)
|
||
|
|
@
|
||
|
|
|
||
|
|
\subsection{Sideways tables}
|
||
|
|
Requires \verb|\usepackage{rotating}| in the \LaTeX\
|
||
|
|
preamble. Sideways tables can't be forced in place with the \code{[H]}
|
||
|
|
specifier, but you can use the \verb|\clearpage| command to get them
|
||
|
|
fairly nearby.
|
||
|
|
|
||
|
|
<<results='asis'>>=
|
||
|
|
x <- x[1:30, ]
|
||
|
|
x.side <- xtable(x, caption = "A sideways table")
|
||
|
|
print(x.side, floating = TRUE, floating.environment = "sidewaystable")
|
||
|
|
@
|
||
|
|
\clearpage
|
||
|
|
|
||
|
|
\subsection{Rescaled tables}
|
||
|
|
Specify a \code{scalebox} value to rescale the table.
|
||
|
|
<<results='asis'>>=
|
||
|
|
x <- x[1:20, ]
|
||
|
|
x.rescale <- xtable(x)
|
||
|
|
print(x.rescale, scalebox = 0.7)
|
||
|
|
@
|
||
|
|
|
||
|
|
\subsection{Aligning fixed width columns}
|
||
|
|
Note that using specifications such as \verb|p{2cm}| always
|
||
|
|
produces a \textbf{left aligned} column. What if some other alignment
|
||
|
|
is desired?
|
||
|
|
|
||
|
|
This is not really a problem with \pkg{xtable} but with the formatting
|
||
|
|
of tables with fixed width columns and different alignments using
|
||
|
|
standard \LaTeX.
|
||
|
|
|
||
|
|
One solution is to use the \verb|array| package, defining new
|
||
|
|
column formats.
|
||
|
|
|
||
|
|
\begin{verbatim}
|
||
|
|
\newcolumntype{L}[1]{>{\raggedright\let\newline\\
|
||
|
|
\arraybackslash\hspace{0pt}}m{#1}}
|
||
|
|
\newcolumntype{C}[1]{>{\centering\let\newline\\
|
||
|
|
\arraybackslash\hspace{0pt}}m{#1}}
|
||
|
|
\newcolumntype{R}[1]{>{\raggedleft\let\newline\\
|
||
|
|
\arraybackslash\hspace{0pt}}m{#1}}
|
||
|
|
\newcolumntype{P}[1]{>{\raggedright\tabularxbackslash}p{#1}}
|
||
|
|
\end{verbatim}
|
||
|
|
|
||
|
|
These allow for very sophisticated cell formatting, namely
|
||
|
|
left-aligned, centred, or right-aligned text, with recognition of line
|
||
|
|
breaks for the first three new column types. If these lines are
|
||
|
|
included along with \verb|\usepackage{array}|, then the following is
|
||
|
|
possible.
|
||
|
|
|
||
|
|
\newcolumntype{L}[1]{>{\raggedright\let\newline\\
|
||
|
|
\arraybackslash\hspace{0pt}}m{#1}}
|
||
|
|
\newcolumntype{C}[1]{>{\centering\let\newline\\
|
||
|
|
\arraybackslash\hspace{0pt}}m{#1}}
|
||
|
|
\newcolumntype{R}[1]{>{\raggedleft\let\newline\\
|
||
|
|
\arraybackslash\hspace{0pt}}m{#1}}
|
||
|
|
\newcolumntype{P}[1]{>{\raggedright\tabularxbackslash}p{#1}}
|
||
|
|
|
||
|
|
<<results='asis'>>=
|
||
|
|
df <- data.frame(name = c("A","B"), right = c(1.4, 34.6),
|
||
|
|
left = c(1.4, 34.6), text = c("txt1","txt2"))
|
||
|
|
print(xtable(df, align = c("l", "|c", "|R{3cm}", "|L{3cm}", "| p{3cm}|")),
|
||
|
|
floating = FALSE, include.rownames = FALSE)
|
||
|
|
@
|
||
|
|
|
||
|
|
\newpage
|
||
|
|
|
||
|
|
\subsection{Table width}
|
||
|
|
The \code{tabularx} environment is for typesetting tables whose
|
||
|
|
overall width is fixed. The column alignment code \code{X} denotes
|
||
|
|
columns that will be stretched to achieve the desired table
|
||
|
|
width. Requires \verb|\usepackage{tabularx}| in the \LaTeX\ preamble.
|
||
|
|
|
||
|
|
<<results='asis'>>=
|
||
|
|
df.width <- data.frame(One = c("item 1", "A"), Two = c("item 2", "B"),
|
||
|
|
Three = c("item 3", "C"), Four = c("item 4", "D"))
|
||
|
|
x.width <- xtable(df.width)
|
||
|
|
align(x.width) <- "|l|X|l|l|l|"
|
||
|
|
print(x.width, tabular.environment = "tabularx", width = "\\textwidth")
|
||
|
|
@
|
||
|
|
|
||
|
|
\section{Suppressing printing}
|
||
|
|
By default the \code{print} method will print the \LaTeX\ or HTML to
|
||
|
|
standard output and also return the character strings invisibly. The
|
||
|
|
printing to standard output can be suppressed by specifying
|
||
|
|
\code{print.results = FALSE}.
|
||
|
|
|
||
|
|
<<>>=
|
||
|
|
x.out <- print(tli.table, print.results = FALSE)
|
||
|
|
@
|
||
|
|
|
||
|
|
Formatted output can also be captured without printing with the
|
||
|
|
\code{toLatex} method. This function returns an object of class
|
||
|
|
\code{"Latex"}.
|
||
|
|
|
||
|
|
<<>>=
|
||
|
|
x.ltx <- toLatex(tli.table)
|
||
|
|
class(x.ltx)
|
||
|
|
x.ltx
|
||
|
|
@
|
||
|
|
|
||
|
|
|
||
|
|
\newpage
|
||
|
|
|
||
|
|
\section{Acknowledgements}
|
||
|
|
Most of the examples in this gallery are taken from the \pkg{xtable}
|
||
|
|
documentation. Two examples (\code{add.to.row} and `Aligning fixed width
|
||
|
|
columns') are from Stack Exchange.
|
||
|
|
|
||
|
|
\section{Session information}
|
||
|
|
<<results='asis'>>=
|
||
|
|
toLatex(sessionInfo())
|
||
|
|
@
|
||
|
|
|
||
|
|
\end{document}
|