Adjust decimals on table from csv

Question

My problem is that I'm importing some data from a csv and generating a table.

The table is good, but I need the numbers with 4 decimals. I've generated the table with this code (probably it is not the cleanest one, but it works):

\documentclass{article}

\usepackage{booktabs}
\usepackage{csvsimple}

\newcommand{\dataa}{\begin{tabular}{c} \\  \\ \\ Var1 \\ \\  \\ \\Var2 \\  \end{tabular}}
\newcommand{\datab}{\begin{tabular}{c}  4\\ 8\\12\\16\\4\\ 8\\12\\16\\  \end{tabular}}
\newcommand{\datac}{\csvreader[separator=comma, no head, tabular={c},filter not strcmp={\boot}{}]{Tab_2.csv}{5=\expa,6=\rec,7=\dri,8=\boot}{\expa}}
\newcommand{\datad}{\csvreader[separator=comma, no head, tabular={c},filter not strcmp={\boot}{}]{Tab_2.csv}{5=\expa,6=\rec,7=\dri,8=\boot}{\rec}}
\newcommand{\datae}{\csvreader[separator=comma, no head, tabular={c},filter not strcmp={\boot}{}]{Tab_2.csv}{5=\expa,6=\rec,7=\dri,8=\boot}{\dri}}
\newcommand{\dataf}{\csvreader[separator=comma, no head, tabular={c},filter not strcmp={\boot}{}]{Tab_2.csv}{5=\expa,6=\rec,7=\dri,8=\boot}{\boot}}

\begin{document}
\begin{table}
\centering
\begin{tabular}{ l c c c c c} 
\hline \hline
& & \multicolumn{2}{c}{Regime} & \multicolumn{2}{c}{Significance} \\
\cmidrule(lr){3-4} \cmidrule(l){5-6} 
Impact & Horizon & State1 & State2 & CI & Bootstrap\\
\hline
\dataa & \datab & \datac & \datad & \datae & \dataf \\
\hline
\end{tabular}
\end{table}

\end{document}

The csv file I'm using has this format (I can't change it, it's been handled to me like this):

,,,,,,,
,,,,,,,
,,,,,,,
,,,,-0.010782,-0.010116,0.48334,0.1108
,,,,-0.041457,-0.015253,0.1942,0.0297
,,,,-0.075416,-0.026916,0.11368,0.0127
,,,,-0.083706,-0.028817,0.11713,0.0259
,,,,0.0018477,-0.0098031,0.20488,0.4541
,,,,-0.0097454,-0.018515,0.38455,0.2622
,,,,-0.025355,-0.0038709,0.2999,0.2734
,,,,-0.042006,0.014067,0.1024,0.283

I have a look for similar problems, but none helped me, I've also been through the help file of the package csvsimple without luck. I also apologize if this issue has already been addressed.

Answer 1

The following code utilizes siunitx. It is based on this answer showing how to establish compatibility between siunitx and csvsimple.

\documentclass{article}

\usepackage{booktabs}
\usepackage{csvsimple}
\usepackage[round-mode=places, round-integer-to-decimal, round-precision=4,
    table-format = 1.4, 
    table-number-alignment=center,
    round-integer-to-decimal]{siunitx}
\newcolumntype{N}{c@{}S}

\newcommand{\dataa}{\begin{tabular}{c} \\  \\ \\ Var1 \\ \\  \\ \\Var2 \\  \end{tabular}}
\newcommand{\datab}{\begin{tabular}{c}  4\\ 8\\12\\16\\4\\ 8\\12\\16\\  \end{tabular}}
\newcommand{\datac}{\csvreader[separator=comma, no head, tabular={N},filter not strcmp={\boot}{}]{test.csv}{5=\expa,6=\rec,7=\dri,8=\boot}{& \expa}}
\newcommand{\datad}{\csvreader[separator=comma, no head, tabular={N},filter not strcmp={\boot}{}]{test.csv}{5=\expa,6=\rec,7=\dri,8=\boot}{& \rec}}
\newcommand{\datae}{\csvreader[separator=comma, no head, tabular={N},filter not strcmp={\boot}{}]{test.csv}{5=\expa,6=\rec,7=\dri,8=\boot}{& \dri}}
\newcommand{\dataf}{\csvreader[separator=comma, no head, tabular={N},filter not strcmp={\boot}{}]{test.csv}{5=\expa,6=\rec,7=\dri,8=\boot}{& \boot}}

\begin{document}
    \begin{table}
        \centering
        \begin{tabular}{ l c c c c c } 
            \toprule
            & & \multicolumn{2}{c}{Regime} & \multicolumn{2}{c}{Significance}\\\cmidrule(lr){3-4} \cmidrule(l){5-6} 
            Impact & Horizon & State1 & State2 & CI & Bootstrap\\\midrule
            \dataa & \datab & \datac & \datad & \datae & \dataf \\\bottomrule
        \end{tabular}
    \end{table}
\end{document}