# Aligning table notes to the left in three part table

I have some trouble with the table notes of my table. For some reason, my table notes do not align to the left (probably because my table is quite wide). Is there a way to widen the table notes such that it fits the table? Or make the table smaller such that it fits my table notes?

\documentclass{article}
\usepackage[flushleft]{threeparttable}
\usepackage{booktabs}
\begin{document}
\begin{table}[htb!]
\caption{\textbf{Estimates of the MS-DNS model}}
\label{table:msdns_estimation_results}
\begin{threeparttable}
\footnotesize
\renewcommand{\TPTminimum}{\linewidth}
\makebox[\linewidth]{%
\tabcolsep=0.11cm
\begin{tabular}{lccc|ccc}
Panel (A): Constant and autogressive coefficients of AR  \\
\toprule
\hline
& \thead[l]{$\mathrm{Level}_{t-1}$} & \thead[l]{$\mathrm{Slope}_{t-1}$} &
\thead[l]{$\mathrm{Curvature}_{t-1}$} & $\boldsymbol{\mu}_1$ &
$\boldsymbol{\mu}_2$ \\
\midrule
$\mathrm{Level}_{t}$ $(\beta_{1,t})$ & \bm{0.991}  &    &   &  0.052 & \\
& (0.006)     &    & & (0.036) & \\
$\mathrm{Slope}_{t}$ $(\beta_{2,t})$ &    &    &     & \bm{-1.190} &
\bm{-4.503} \\
&      &               &  & (0.089) & (0.107) \\
$\mathrm{Curvature}_{t}$ $(\beta_{3,t})$ &  & & \bm{0.944} & -0.086 & \\
&      &               & (0.017) & (0.047) & \\
\hline
\\
Panel (B): Variance matrix of AR and decay parameter $\lambda$  \\
\toprule
\hline
& \thead[l]{$\mathrm{Level}_{t}$ $(\beta_{1,t})$} & \thead[l]
{$\mathrm{Slope}_{t}$ $(\beta_{2,t})$} & \thead[l]{$\mathrm{Curvature}_{t}$
$(\beta_{3,t})$} & \lambda & $p_{11}$ & $p_{22}$ \\
\midrule
$\mathrm{Level}_{t}$ $(\beta_{1,t})$ & \bm{0.082}  &    &   & \bm{0.041} &
\bm{0.975} & \bm{0.971} \\
& (0.007)     &               & & (0.001) & (0.011) & (0.012) \\
$\mathrm{Slope}_{t}$ $(\beta_{2,t})$ &    & \bm{1.038}   &     & & & \\
&      &  (0.078)          & & & & \\
$\mathrm{Curvature}_{t}$ $(\beta_{3,t})$ &  & & \bm{0.622} & & &\\
&      &               & (0.066) &  &   & \\
\bottomrule
\end{tabular}
}
\begin{tablenotes}[flushleft]\footnotesize
\item Note: This table reports estimates of the MS-DNS model where the
latent slope and curvature factor follow an AR process. Panel (A) presents
estimates for the autoregressive coefficient matrix $\boldsymbol{F}$ and
vector of means during periods of normal interest rates
($\boldsymbol{\mu}_1$) and in a low interest rate environment
($\boldsymbol{\mu}_2$). Panel (B) presents estimates of the covariance
matrix $\boldsymbol{\Sigma}_{\eta}$, decay parameter $\lambda$, and the
transition probabilities. Bold entries denote parameter estimates
significant at the 5 percent level. Standard errors appear in parentheses.
\par
\end{tablenotes}
\end{threeparttable}
\end{table}
\end{document}


For the interested, my table looks like this:

As it is, your table overflows into the left margin., which explains why the table notes do not look left-aligned. You should have used a \multicolumn for the ‘Panel’ rows. Other than that, the \bm command requires to be in math mode. Also, don't use vertical rules with booktabs, or neutralise the vertical padding around it rules, as they normally do not intersect.

I loaded caption to have a better skip between caption and table, and added various improvements.

    \documentclass{article}
\usepackage{amssymb, amsmath, bm}
\usepackage[flushleft]{threeparttable}
\usepackage{makecell, caption}
\usepackage{booktabs}

\begin{document}

\begin{table}[htb!]
\caption{\textbf{Estimates of the MS-DNS model}}
\label{table:msdns_estimation_results}
\begin{threeparttable}
\footnotesize
\renewcommand{\TPTminimum}{\linewidth}
\makebox[\linewidth]{%
\tabcolsep=0.11cm
\setlength{\extrarowheight}{2pt}
%\setcellgapes{3pt}\makegapedcells
\begin{tabular}{lccc|ccc}
\multicolumn{7}{l}{Panel (A): Constant and autogressive coefficients of AR} \\
\bottomrule
\specialrule{0.4pt}{\aboverulesep}{0pt}
& \thead[l]{$\mathrm{Level}_{t-1}$} & \thead[l]{$\mathrm{Slope}_{t-1}$} &
\thead[l]{$\mathrm{Curvature}_{t-1}$} & $\boldsymbol{\mu}_1$ &
$\boldsymbol{\mu}_2$ \\
\Xhline{0.5pt}
$\mathrm{Level}_{t}$ $(\beta_{1,t})$ & $\bm{0.991}$ & & & 0.052 & \\
& (0.006) & & & (0.036) & \\
$\mathrm{Slope}_{t}$ $(\beta_{2,t})$ & & & & $\bm{-1.190}$ &
$\bm{-4.503}$ \\
& & & & (0.089) & (0.107) \\
$\mathrm{Curvature}_{t}$ $(\beta_{3,t})$ & & & $\bm{0.944}$ & -0.086 & \\
& & & (0.017) & (0.047) & \\
\Xhline{0.8pt}
\\
\multicolumn{7}{l}{ Panel (B): Variance matrix of AR and decay parameter $\lambda$} \\
\bottomrule
\specialrule{0.4pt}{\aboverulesep}{0pt}
& \thead[l]{$\mathrm{Level}_{t}$ $(\beta_{1,t})$} & \thead[l]
{$\mathrm{Slope}_{t}$ $(\beta_{2,t})$} & \thead[l]{$\mathrm{Curvature}_{t}$
$(\beta_{3,t})$} & $\lambda$ & $p_{11}$ & $p_{22}$ \\
\Xhline{0.5pt}
$\mathrm{Level}_{t}$ $(\beta_{1,t})$ & $\bm{0.082}$ & & & $\bm{0.041}$ &
$\bm{0.975}$ & $\bm{0.971}$ \\
& (0.007) & & & (0.001) & (0.011) & (0.012) \\
$\mathrm{Slope}_{t}$ $(\beta_{2,t})$ & & $\bm{1.038}$ & & & & \\
& & (0.078) & & & & \\
$\mathrm{Curvature}_{t}$ $(\beta_{3,t})$ & & & $\bm{0.622}$ & & &\\
& & & (0.066) & & & \\
\Xhline{0.8pt}
\end{tabular}
}
\begin{tablenotes}[flushleft]\footnotesize\smallskip
\item Note: This table reports estimates of the MS-DNS model where the
latent slope and curvature factor follow an AR process. Panel (A) presents
estimates for the autoregressive coefficient matrix $\boldsymbol{F}$ and
vector of means during periods of normal interest rates
($\boldsymbol{\mu}_1$) and in a low interest rate environment
($\boldsymbol{\mu}_2$). Panel (B) presents estimates of the covariance
matrix $\boldsymbol{\Sigma}_{\eta}$, decay parameter $\lambda$, and the
transition probabilities. Bold entries denote parameter estimates
significant at the 5 percent level. Standard errors appear in parentheses.
\par
\end{tablenotes}
\end{threeparttable}
\end{table}

\end{document}


• Thank you very much kind stranger. – Matthew May 26 at 14:04

With siunitx (for aligning numbers at decimal points), booktabs for table rules, makecell (for column headers in two lines), threeparttablex (for table notes with command \notes) and caption for caption set up. Used is \normalsize font size:

\documentclass{article}
\usepackage{bm}
\usepackage[referable]{threeparttablex}
\usepackage{booktabs, makecell}
\usepackage[labelfont=bf, skip=1ex]{caption}
\usepackage{siunitx}
\usepackage{etoolbox}
\newrobustcmd{\B}{\bfseries}    % <-- for indicate boldface numbers

\begin{document}
\begin{table}[htb!]
\caption{Estimates of the MS-DNS model}
\label{table:msdns_estimation_results}
\begin{threeparttable}
\small
\setlength\tabcolsep{0pt}
\begin{tabular*}{\linewidth}{@{\extracolsep{\fill}}
l*{6}{S[input-symbols = {( - )},
detect-weight,
mode=text,
table-format=-1.3]}
}
\multicolumn{7}{l}{Panel (A): Constant and autogressive coefficients of AR}  \\
\toprule
& {Level$_{t-1}$}
& {Slope$_{t-1}$}
& {Curvature$_{t-1}$}
& {$\bm{\mu}_1$}
& {$\bm{\mu}_2$} \\
\midrule
Level$_{t}\ (\beta_{1,t})$
& \B 0.991  &   &           &  0.052    &               \\
& (0.006)   &   &           & (0.036)   &               \\
Slope$_{t}\ (\beta_{2,t})$
&           &   &           & \B -1.190 & \B -4.503     \\
&           &   &           & (0.089)   & (0.107)       \\
Curvature$_{t}\ (\beta_{3,t})$
&           &   & \B 0.944  & -0.086    &               \\
&               &   & (0.017)
& (0.047)   &               \\
\bottomrule
\multicolumn{7}{l}{Panel (B): Variance matrix of AR and decay parameter $\lambda$}  \\
\toprule
& {\makecell[b]{Level$_{t}$\\ $(\beta_{1,t})$}}
& {\makecell[b]{Slope$_{t}$\\ $(\beta_{2,t})$}}
& {\makecell[b]{Curvature$_{t}$\\ $(\beta_{3,t})$}}
& {$\lambda$}
& {$p_{11}$}
& {$p_{22}$}                            \\
\midrule
Level$_{t}\ (\beta_{1,t})$
& \B 0.082  &   &   & \B 0.041  & \B 0.975  & \B 0.971      \\
& (0.007)   &   &   & (0.001)   & (0.011)   & (0.012)       \\
Slope$_{t}\ (\beta_{2,t})$
&           & \B 1.038
&   &           &           &               \\
&           &(0.078)
&   &           &           &               \\
Curvature$_{t}\ (\beta_{3,t})$
&           &   & \B 0.622
&           &           &               \\
&           &   & (0.066)
&           &           &               \\
\bottomrule
\end{tabular*}
\begin{tablenotes}[flushleft]\footnotesize
\note{This table reports estimates of the MS-DNS model where the
latent slope and curvature factor follow an AR process. Panel (A) presents
estimates for the autoregressive coefficient matrix $\bm{F}$ and
vector of means during periods of normal interest rates
($\bm{\mu}_1$) and in a low interest rate environment
($\bm{\mu}_2$). Panel (B) presents estimates of the covariance
matrix $\bm{\Sigma}_{\eta}$, decay parameter $\lambda$, and the
transition probabilities. Bold entries denote parameter estimates
significant at the 5 percent level. Standard errors appear in parentheses.}
\end{tablenotes}
\end{threeparttable}
\end{table}
\end{document}


(red lines indicate text borders)

• Thank you for your help – Matthew May 26 at 14:05