I am trying to do something quite similar to this post:
pgfplotstable: Conditional post-processing of cell content on a per-column basis
However, I'd like to post-process the column based on the content from another column. For example, I want to add a * to the coefficient column if the corresponding t-value is greater than 2.
In the table, usually I only report the coefficient and standard error, not the t-value. I'd like to post-process the coefficientcolumns based on values from coefficient/standard error.
Below is the MWE:
\documentclass{article}
\usepackage{pgfplotstable}
\begin{filecontents*}{test1.dat}
coefficient {standard error}
-.0159375 .008852
-.0107286 .0091658
.0042201 .0089453
.0108719 .0038041
\end{filecontents*}
\begin{document}
\pgfplotstableread{test1.dat}\results
\pgfplotstablecreatecol[expr={\thisrow{coefficient}/\thisrow{standard error}}]{t-value}{\results}
\def\bordervalue{2}
\pgfplotstabletypeset[
columns={coefficient, standard error, t-value},
columns/t-value/.style={
%preproc/expr = {100*##1},
postproc cell content/.style={
/pgfplots/table/@cell content/.add={}{%
\pgfmathparse{int(greater(##1,\bordervalue))}
\ifnum\pgfmathresult=1
$^*$
\fi
},
},
},
]\results
\end{document}

How can I put the * to the coefficient column so that I don't need to report the t-value?
Any help is appreciated.
