Ich bin ein Student im letzten Jahr der Politikwissenschaft, aber ein Anfänger im LaTeX-Satz. Bitte verzeihen Sie mir, wenn ich dumm klinge.
Ich formatierte eine Tabelle für die Darstellung meiner Erkenntnisse aus einer von mir durchgeführten Regressionsanalyse. Ich brauche bitte Hilfe beim Ausrichten der Beschreibungszellen mit den dargestellten Ergebnissen.
Hier ist beispielsweise die Tabellenausgabe, die ich gerade habe:
Wie Sie sehen, sind beispielsweise die Zellen in den Spalten 2, 3 und 4 nicht korrekt mit der Beschreibung in Spalte 1 ausgerichtet. Es gibt eine starke Verzerrung nach oben und unten, die ich unbedingt korrigieren möchte.
Hier ist der Code, den ich derzeit verwende. Ich wäre sehr dankbar, wenn mir jemand den Code selbst erklären könnte.
\begin{tabular}{@{}llll@{}}
\multicolumn{4}{c}{\textbf{\begin{tabular}[c]{@{}c@{}}Table 3. OLS (Ordinary Least Square) Regression Analysis Between \\ Development Indicators\\ And Infant Mortality Rate In India\end{tabular}}} \\ \midrule
\multicolumn{1}{c}{\multirow{2}{*}{Independent Variables}} & \multicolumn{3}{c}{Infant Mortality Rate} \\ \cmidrule(l){2-4}
\multicolumn{1}{c}{} & \multicolumn{1}{c}{Model 1} & \multicolumn{1}{c}{Model 2} & \multicolumn{1}{c}{Model 3} \\ \midrule
\textit{1. Health Inputs, Access, and Participation} & & & \\
Percent Mothers Receiving Antenatal Care & \begin{tabular}[c]{@{}l@{}}-0.2568*\\ (0.1146)\end{tabular} & \begin{tabular}[c]{@{}l@{}}-0.2474*\\ (0.0896)\end{tabular} & \begin{tabular}[c]{@{}l@{}}-0.2969**\\ (0.0851)\end{tabular} \\
Percent Mothers Receiving JSY & \begin{tabular}[c]{@{}l@{}}0.3005**\\ (0.0860)\end{tabular} & \begin{tabular}[c]{@{}l@{}}0.2952***\\ (0.0735)\end{tabular} & \begin{tabular}[c]{@{}l@{}}0.3004***\\ (0.0751)\end{tabular} \\
Avg. Delivery Expenditure (Rs.) & \begin{tabular}[c]{@{}l@{}}-0.0018*\\ (0.0008)\end{tabular} & \begin{tabular}[c]{@{}l@{}}-0.0018*\\ (0.0008)\end{tabular} & \begin{tabular}[c]{@{}l@{}}-0.0024**\\ (0.0007)\end{tabular} \\
\textit{2. Social Indicators} & & & \\
Percent Adult Females Literate & \begin{tabular}[c]{@{}l@{}}-0.0154\\ (0.2128)\end{tabular} & \multicolumn{1}{c}{—} & \multicolumn{1}{c}{—} \\
Percent Children Underweight & \begin{tabular}[c]{@{}l@{}}0.2440\\ (0.2318)\end{tabular} & \begin{tabular}[c]{@{}l@{}}0.2468\\ (0.1655)\end{tabular} & \multicolumn{1}{c}{—} \\
\textit{3. Modernization Indicators} & & & \\
NSDP Per Capita (Rs.) & \begin{tabular}[c]{@{}l@{}}0.73e-05\\ (0.3e-04)\end{tabular} & \multicolumn{1}{c}{—} & \multicolumn{1}{c}{—} \\
Percent Households using Water Source & \begin{tabular}[c]{@{}l@{}}0.0114\\ (0.1912)\end{tabular} & \multicolumn{1}{c}{—} & \multicolumn{1}{c}{—} \\ \midrule
F-statistic & 10.91*** & 21.60*** & 26.80*** \\
R-squared & 0.7763 & 0.7756 & 0.7557 \\
Intercept Estimate & 28.80 & 29.37 & 39.56 \\
N & 30 & 30 & 30 \\ \midrule
\multicolumn{4}{l}{\begin{tabular}[c]{@{}l@{}}Note: Coefficients are unstandardized and include standard error in parentheses\\ \\ * p \textless 0.05\\ ** p \textless 0.01\\ *** p \textless 0.001\end{tabular}} \\ \bottomrule
\end{tabular}
Nochmals vielen Dank!!
Antwort1
\documentclass{article}
\usepackage{booktabs}
\usepackage{multirow}
\begin{document}
\begin{table}
\caption{OLS (Ordinary Least Square) Regression Analysis Between Development Indicators And Infant Mortality Rate In India}
\label{key}
\setlength{\tabcolsep}{0pt}
\begin{tabular*}{\textwidth}{@{\extracolsep{\fill}}llll}
\toprule
\multicolumn{1}{c}{\multirow{2}{*}{Independent Variables}} & \multicolumn{3}{c}{Infant Mortality Rate} \\
\cmidrule(l){2-4}
& \multicolumn{1}{c}{Model 1} & \multicolumn{1}{c}{Model 2} & \multicolumn{1}{c}{Model 3} \\
\midrule
\textit{1. Health Inputs, Access, and Participation} \\
Percent Mothers Receiving Antenatal Care & -0.2568* & -0.2474* & -0.2969** \\
& (0.1146) & (0.0896) & (0.0851) \\
Percent Mothers Receiving JSY & 0.3005** & 0.2952*** & 0.3004*** \\
& (0.0860) & (0.0735) & (0.0751) \\
Avg. Delivery Expenditure (Rs.) & -0.0018* & -0.0018* & -0.0024** \\
& (0.0008) & (0.0008) & (0.0007) \\
\textit{2. Social Indicators} \\
Percent Adult Females Literate & -0.0154 & \multicolumn{1}{c}{—} & \multicolumn{1}{c}{—} \\
& (0.2128) & \\
Percent Children Underweight & 0.2440 & 0.2468 & \multicolumn{1}{c}{—} \\
& (0.2318) & (0.1655) \\
\textit{3. Modernization Indicators} \\
NSDP Per Capita (Rs.) & 0.73e-05 & \multicolumn{1}{c}{—} & \multicolumn{1}{c}{—} \\
& (0.3e-04) \\
Percent Households using Water Source & 0.0114 & \multicolumn{1}{c}{—} & \multicolumn{1}{c}{—} \\
& (0.1912) \\
\midrule
F-statistic & 10.91*** & 21.60*** & 26.80*** \\
R-squared & 0.7763 & 0.7756 & 0.7557 \\
Intercept Estimate & 28.80 & 29.37 & 39.56 \\
N & 30 & 30 & 30 \\
\midrule
\end{tabular*}
Note: Coefficients are unstandardized and include standard error in parentheses
* p \textless 0.05 ** p \textless 0.01 *** p \textless 0.001
\end{table}
\end{document}
Antwort2
Ich würde das dcolumn
Paket verwenden, um die Zahlen in den drei Datenspalten an ihren Dezimalstellen auszurichten. Außerdem würde ich den \caption
Befehl verwenden, um die Überschrift zu setzen, und ich würde eine threeparttable
Umgebung verwenden, um die Breite der Überschrift auf die des tabular
Materials einzustellen.
\documentclass{article}
\usepackage[margin=2.5cm]{geometry} % set page parameters appropriately
\usepackage{booktabs,threeparttable}
\usepackage[font=bf,skip=0.5\baselineskip]{caption}
\usepackage{dcolumn}
\newcolumntype{d}[1]{D..{#1}}
\newcommand\mc[1]{\multicolumn{1}{c}{#1}} % handy shortcut macro
\begin{document}
\addtocounter{table}{2}
\begin{table}[ht!]
\centering
\begin{threeparttable}
\caption{OLS (Ordinary Least Square) Regression Analysis between Development Indicators and Infant Mortality Rate in India}
\begin{tabular}{@{}l*{3}{d{2.6}}@{}}
\toprule
Independent Variables &
\multicolumn{3}{c@{}}{Infant Mortality Rate} \\
\cmidrule(l){2-4}
& \mc{Model 1} & \mc{Model 2} & \mc{Model 3} \\
\midrule
\em 1. Health Inputs, Access, and Participation \\
Percent Mothers Receiving Antenatal Care
& -0.2568^{*} & -0.2474^{*} & -0.2969^{**} \\
& (0.1146) & (0.0896) & (0.0851) \\
Percent Mothers Receiving JSY
& 0.3005^{**} & 0.2952^{***} & 0.3004^{***} \\
& (0.0860) & (0.0735) & (0.0751) \\
Avg. Delivery Expenditure (Rs.)
& -0.0018^{*} & -0.0018^{*} & -0.0024^{**} \\
& (0.0008) & (0.0008) & (0.0007) \\
\addlinespace
\em 2. Social Indicators \\
Percent Adult Females Literate
& -0.0154 & \mc{—} & \mc{—} \\
& (0.2128) \\
Percent Children Underweight
& 0.2440 & 0.2468 & \mc{—} \\
& (0.2318) & (0.1655) \\
\addlinespace
\em 3. Modernization Indicators \\
NSDP Per Capita (Rs.)
& \mc{0.73e-05} & \mc{—} & \mc{—} \\
& \mc{(0.3e-04)} \\
Percent Households using Water Source
& 0.0114 & \mc{—} & \mc{—} \\
& (0.1912) \\
\midrule
$F$-statistic & 10.91^{***} & 21.60^{***} & 26.80^{***} \\
$R^2$ & 0.7763 & 0.7756 & 0.7557 \\
Intercept Estimate & 28.80 & 29.37 & 39.56 \\
$N$ & 30 & 30 & 30 \\
\midrule
\multicolumn{4}{@{}l@{}}{Note: Coefficients are unstandardized; standard errors shown in parentheses.}\\
\multicolumn{4}{@{}l@{}}{$^{*}\ p<0.05$; $^{**}\ p<0.01$; $^{***}\ p<0.001$.} \\
\bottomrule
\end{tabular}
\end{threeparttable}
\end{table}
\end{document}
Antwort3
Mit Nutzung aller threetablex
Features, siunitx
Paketierung für S
Spalten, korrekter LaTeX-Syntax und konsistenter Verwendung des Et-Zeichens in der Tabelle:
\documentclass{article}
\usepackage{booktabs}
\usepackage[referable]{threeparttablex}
\usepackage[font={small,bf},
skip=0.33\baselineskip]{caption}
\usepackage{siunitx}
\usepackage{xparse}
\NewExpandableDocumentCommand\mcc{O{1}m}
{\multicolumn{#1}{c}{#2}}
\addtocounter{table}{2}
\begin{document}
\begin{table}[ht!]
\sisetup{input-symbols = {( )},
table-align-text-pre=false,
table-space-text-post={$^{***}$},
table-align-text-post=false,
table-format=-1.4,
group-digits = false,
output-exponent-marker={e},
tight-spacing=true
}
\setlength\tabcolsep{1.5pt}
\begin{threeparttable}
\caption{OLS (Ordinary Least Square) Regression Analysis between Development Indicators and Infant Mortality Rate in India}
\label{tab:regresion}
\begin{tabular}{@{} l SSS @{}}
\toprule
Independent Variables & \mcc[3]{Infant Mortality Rate} \\
\cmidrule(l){2-4}
& \mcc{Model 1} & \mcc{Model 2} & \mcc{Model 3} \\
\midrule
\emph{1. Health Inputs, Access, and Participation} &&& \\
Percent Mothers Receiving Antenatal Care
& -0.2568\tnote{*} & -0.2474\tnote{*} & -0.2969\tnote{**} \\
& (0.1146) & (0.0896) & (0.0851) \\
Percent Mothers Receiving JSY
& 0.3005\tnote{**} & 0.2952\tnote{***} & 0.3004\tnote{***} \\
& (0.0860) & (0.0735) & (0.0751) \\
Avg. Delivery Expenditure (Rs.)
& -0.0018\tnote{*} & -0.0018\tnote{*} & -0.0024\tnote{**} \\
& (0.0008) & (0.0008) & (0.0007) \\
\addlinespace
\emph{1. Social Indicators} &&& \\
Percent Adult Females Literate
& -0.0154 & {--} & {--} \\
& (0.2128) & {--} & {--} \\
Percent Children Underweight
& 0.2440 & 0.2468 & {--} \\
& (0.2318) & (0.1655) & {--} \\
\addlinespace
\emph{3. Modernization Indicators} &&& \\
NSDP Per Capita (Rs.)
& {\num{0.73e-5}} & {--} & {--} \\
& {(\num{0.30e-4})} & {--} & {--} \\
Percent Households using Water Source
& 0.0114 & {--} & {--} \\
& (0.1912) & {--} & {--} \\
\midrule
$F$-statistic
& 10.91\tnote{***} & 21.60\tnote{***} & 26.80\tnote{***} \\
$R^2$
& 0.7763 & 0.7756 & 0.7557 \\
Intercept Estimate
& 28.80 & 29.37 & 39.56 \\
$N$ & 30 & 30 & 30 \\
\bottomrule
\end{tabular}\footnotesize
\begin{tablenotes}[flushleft, para]
\note{Coefficients are unstandardized; standard errors shown in parentheses.
\item[*]: $p<0.05$;
\item[**]: $p<0.01$;
\item[***]: $p<0.001$.}
\end{tablenotes}
\end{threeparttable}
\end{table}
\end{document}
Antwort4
Ich schlage vor, Ihre Tabelle mit siunitx
und zu erstellen threeparttable
:
\documentclass{article}
\usepackage{multirow}
\usepackage{booktabs, caption}
\usepackage{threeparttable}
\usepackage{siunitx}
\begin{document}
\begin{threeparttable}
\setcounter{table}{2}\captionsetup{font=bf, format=hang, labelsep=period}
\setlength{\tabcolsep}{3pt}
\sisetup{table-format=-1.4, table-space-text-post=***, table-align-text-post=false, input-symbols={()}, group-digits=false}
\caption{OLS (Ordinary Least Square) Regression Analysis Between \\ Development Indicators And Infant Mortality Rate In India}
\centering\small
\begin{tabular}{@{}lSSS@{}}\toprule
\multirow{2}{*}{\centering Independent Variables} & \multicolumn{3}{c}{Infant Mortality Rate} \\ \cmidrule(l){2-4}
& {Model 1} & {Model 2} & {Model 3} \\ \midrule
1. \textit{Health Inputs, Access, and Participation}\\
Percent Mothers Receiving Antenatal Care & 0.2568* & -0.2474 & -0.2969** \\
& (0.1146) & (0.0896) & (0.0851) \\ \addlinespace
Percent Mothers Receiving JSY & 0.3005** & 0.2952*** & 0.3004***\\
& (0.0860) & (0.0735) & (0.0751) \\ \addlinespace
Avg. Delivery Expenditure (Rs.) & -0.0018* & -0.0018* & -0.0024** \\
(0.0008) & (0.0008) & (0.0007) \\ \addlinespace
2. \textit{Social Indicators} \\
Percent Adult Females Literate & -0.0154 & {—} & {—} \\
& (0.2128) \\\addlinespace
Percent Children Underweight & 0.2440 & 0.2468 & {—} \\
& (0.2318) & (0.1655)\\ \addlinespace
3. \textit{Modernization Indicators} \\
NSDP Per Capita (Rs.) & 0.73{$ \,\mathrm{e}-05 $} & \multicolumn{1}{c}{—} & \multicolumn{1}{c}{—} \\
& (0.3{$ \,\mathrm{e}-04 $)} \\
Percent Households using Water Source & 0.0114 & {—} & {—} \\
&(0.1912) \\ \midrule
F-statistic & 10.91*** & 21.60*** & 26.80*** \\
R-squared & 0.7763 & 0.7756 & 0.7557 \\
Intercept Estimate & 28.80 & 29.37 & 39.56 \\
N & 30 & 30 & 30 \\ \bottomrule
\end{tabular}
\begin{tablenotes}[online]
\item[Note:] Coefficients are unstandardized and include standard error in parentheses.
\centering*\enspace$ p < 0.05$,\quad **\enspace$ p < 0.01 $,\quad ***\enspace$ p < 0.001 $.
\end{tablenotes}
\end{threeparttable}
\end{document}