I found this nifty code on github (https://github.com/labreumaia/longtable.stargazer/blob/master/longtable.stargazer.R) that combines stargazer and longtable:
longtable.stargazer = function(..., float = T, longtable.float = F,
longtable.head = T, filename = NULL){
# Capturing stargazer to hack it
require(stargazer)
res = capture.output(
stargazer(..., float = float)
)
# Changing tabulare environment for longtable
res = gsub("tabular", "longtable", res)
# removing floating environment
if(float == T & longtable.float == F){
res[grep("table", res)[1]] = res[grep("longtable", res)[1]]
# Removing extra longtable commands
res = res[-grep("longtable", res)[2]]
res = res[-length(res)]
}
# Adding page headings
if(longtable.head == T){
res = c(res[1:which(res == "\\hline \\\\[-1.8ex] ")[1] - 1], "\\endhead", res[which(res == "\\hline \\\\[-1.8ex] ")[1]:length(res)])
}
# Exporting
cat(res, sep = "\n")
# Exporting
if(!is.null(filename)){
cat(res, file = filename, sep = "\n")
# Message
cat(paste("\nLaTeX output printed to", filename, "\n", sep = " ",
collapse = ""))
}else{
cat(res, sep = "\n")
}
}
But when I conduct the following regression analysis with the cars data:
ols <- lm(speed ~ dist, data = cars)
longtable.stargazer(ols)
I end up outputting two tables. Why is this happening?
This is the output I am getting:
% Table created by stargazer v.5.2.2 by Marek Hlavac, Harvard University. E-mail: hlavac at fas.harvard.edu
% Date and time: Wed, May 05, 2021 - 20:22:44
\begin{table}[!htbp] \centering
\caption{}
\label{}
\begin{longtable}{@{\extracolsep{5pt}}lc}
\\[-1.8ex]\hline
\endhead
\hline \\[-1.8ex]
& \multicolumn{1}{c}{\textit{Dependent variable:}} \\
\cline{2-2}
\\[-1.8ex] & speed \\
\hline \\[-1.8ex]
dist & 0.166$^{***}$ \\
& (0.017) \\
& \\
Constant & 8.284$^{***}$ \\
& (0.874) \\
& \\
\hline \\[-1.8ex]
Observations & 50 \\
R$^{2}$ & 0.651 \\
Adjusted R$^{2}$ & 0.644 \\
Residual Std. Error & 3.156 (df = 48) \\
F Statistic & 89.567$^{***}$ (df = 1; 48) \\
\hline
\hline \\[-1.8ex]
\textit{Note:} & \multicolumn{1}{r}{$^{*}$p$<$0.1; $^{**}$p$<$0.05; $^{***}$p$<$0.01} \\
\end{longtable}
\end{table}
% Table created by stargazer v.5.2.2 by Marek Hlavac, Harvard University. E-mail: hlavac at fas.harvard.edu
% Date and time: Wed, May 05, 2021 - 20:22:44
\begin{table}[!htbp] \centering
\caption{}
\label{}
\begin{longtable}{@{\extracolsep{5pt}}lc}
\\[-1.8ex]\hline
\endhead
\hline \\[-1.8ex]
& \multicolumn{1}{c}{\textit{Dependent variable:}} \\
\cline{2-2}
\\[-1.8ex] & speed \\
\hline \\[-1.8ex]
dist & 0.166$^{***}$ \\
& (0.017) \\
& \\
Constant & 8.284$^{***}$ \\
& (0.874) \\
& \\
\hline \\[-1.8ex]
Observations & 50 \\
R$^{2}$ & 0.651 \\
Adjusted R$^{2}$ & 0.644 \\
Residual Std. Error & 3.156 (df = 48) \\
F Statistic & 89.567$^{***}$ (df = 1; 48) \\
\hline
\hline \\[-1.8ex]
\textit{Note:} & \multicolumn{1}{r}{$^{*}$p$<$0.1; $^{**}$p$<$0.05; $^{***}$p$<$0.01} \\
\end{longtable}
\end{table}
There is a loose
cat()
call, followed by anothercat()
insideif/else
which doubles the output. Comment that line out: