booktabs パッケージを使用して LaTeX で表を作成しましたが、コンパイルするとページ余白を超えてしまいます。表では複数列を使用しています。標準のページ余白内に収まるようにしながら、内容の読みやすさを維持する必要があります。オーバーフローを防ぐために、表またはドキュメントのレイアウトを調整するにはどうすればよいでしょうか。
問題を再現する MWE は次のとおりです。
\documentclass[a4paper, 12pt]{article}
\usepackage[margin=2cm]{geometry}
\usepackage{booktabs}
\usepackage{natbib}
\usepackage{amsmath}
\usepackage{tabulary}
\usepackage{tabularx}
\usepackage{varioref}
\usepackage{multirow}
\begin{document}
\begin{table}[!ht]
\begin{center}
\caption{CIPS Panel Unit Root Tests}
\label{tab:table_CIPS}
\vspace{.5cm}
\begin{tabular}{lcccccc}
\toprule
\toprule
& \multicolumn {2}{c}{Intercept} & \multicolumn {2}{c}{Intercept +\ Trend }&
\multicolumn {2}{c}{Intercept +\ First Difference } \\
\cmidrule(lr){2-3}
\cmidrule(lr){4-5}
\cmidrule(lr){6-7}
&\multicolumn{1}{c}{CADF_{i}(0)}&\multicolumn{1}{c}{CADF_{i}(1)}&\multicolumn{1}{c}
{CADF_{i}(0)}&\multicolumn{1}{c}{CADF_{i}(1)}&\multicolumn{1}{c}{CADF_{i}
(0)}&\multicolumn{1}{c}{CADF_{i}(1)}\\
\cmidrule{2-5}
ln nonresource FDI & -1.86**& 0.92 & -16.23*** &
0.098***\\
ln population & -1.86**& 0.92 & -16.23*** &
0.098***\\
ln human capital & -1.86**& 0.92 & -16.23*** &
0.098***\\
ln GDP per capita $(t = 1)$ & -1.86**& 0.92 & -16.23*** &
0.098***\\
ln GDP surrounding market potential & -1.86**& 0.92 &
-16.23*** & 0.098***\\
Government share of GDP \times 100 & -1.86**& 0.92 & -16.23*** & 0.098***\\
ln hydrocarbon resource rents (t \times 1) & -1.86** & 0.92 & -16.23*** & 0.098***\\
ln nonresource FDI (i \times 1) & -1.86** & 0.92 & -16.23*** & 0.098***\\
\bottomrule
\bottomrule
\end{tabular}%
\end{center}
{\footnotesize
{H0: All series are nonstationary. The statistics are the standardized version of the $CIPS_(p)$ statistic for an unbalanced panel. The $CIPS_(p)$ statistic is the cross-section average of the cross sectionally augmented Dickey-Fuller test statistic $(CADF_{i}(p)))$. Following Pesaran (2007), extreme t values are truncated to avoid any undue influence of extreme outcomes, because t is small (10–20). *$p<0.1$, **$p<0.05$, ***$p<0.01$. For the first difference of ln population, we also reject the null if we restrict the sample to a balanced panel.}}
\end{table}
\end{document}
答え1
横向きにしたらどうですか？
threeparttable表のメモには以下を使用しました:
\documentclass[a4paper, 12pt]{article}
\usepackage[margin=2cm]{geometry}
\usepackage{booktabs}
\usepackage{amsmath}
\usepackage{rotating}
\usepackage[para]{threeparttable}
\usepackage{caption}
\begin{document}
\begin{sidewaystable}\centering
\begin{threeparttable}
\caption{CIPS Panel Unit Root Tests}
\label{tab:table_CIPS}
\begin{tabular}{lcccccc}
\toprule
\toprule
& \multicolumn {2}{c}{Intercept} & \multicolumn {2}{c}{Intercept +\ Trend }&
\multicolumn {2}{c}{Intercept +\ First Difference } \\
\cmidrule(lr){2-3}
\cmidrule(lr){4-5}
\cmidrule(lr){6-7}
&\multicolumn{1}{c}{$\textrm{CADF}_{i}(0)$}&\multicolumn{1}{c}{$\textrm{CADF}_{i}(1)$}&\multicolumn{1}{c}
{$\textrm{CADF}_{i}(0)$}&\multicolumn{1}{c}{$\textrm{CADF}_{i}(1)$}&\multicolumn{1}{c}{$\textrm{CADF}_{i}
(0)$}&\multicolumn{1}{c}{$\textrm{CADF}_{i}(1)$}\\
\cmidrule{2-7}
ln nonresource FDI & -1.86**& 0.92 & -16.23*** &
0.098***\\
ln population & -1.86**& 0.92 & -16.23*** &
0.098***\\
ln human capital & -1.86**& 0.92 & -16.23*** &
0.098***\\
ln GDP per capita $(t = 1)$ & -1.86**& 0.92 & -16.23*** &
0.098***\\
ln GDP surrounding market potential & -1.86**& 0.92 &
-16.23*** & 0.098***\\
Government share of GDP $\times$ 100 & -1.86**& 0.92 & -16.23*** & 0.098***\\
ln hydrocarbon resource rents (t $\times$ 1) & -1.86** & 0.92 & -16.23*** & 0.098***\\
ln nonresource FDI (i $\times$ 1) & -1.86** & 0.92 & -16.23*** & 0.098***\\
\bottomrule
\bottomrule
\end{tabular}
\begin{tablenotes}
\item H0: All series are nonstationary. The statistics are the standardized version of the $CIPS_(p)$ statistic for an unbalanced panel. The $CIPS_(p)$ statistic is the cross-section average of the cross sectionally augmented Dickey-Fuller test statistic $(CADF_{i}(p)))$. Following Pesaran (2007), extreme t values are truncated to avoid any undue influence of extreme outcomes, because t is small (10–20). *$p<0.1$, **$p<0.05$, ***$p<0.01$. For the first difference of ln population, we also reject the null if we restrict the sample to a balanced panel.
\end{tablenotes}
\end{threeparttable}
\end{sidewaystable}
\end{document}
