0

In my research, I usually have to deal with large table of results. I try to find a way to import these results into Latex. I search the internet and try some package like csvsimple, pgfplotstable. At first, these packages work with small table of 3, 4 rows and columns. However, when I add more rows and more columns, the packages won't work anymore. With larger table dimension, there are many errors with floating point number, taking the message below as an example: "! Package PGF Math Error: Could not parse input '(-1.63)' as a floating point number, sorry. The unreadable part was near '(-1.63)'." Does someone have the solution to this problem. I attach an example of what I am working below. The data is in text format, with comma separator. 

\documentclass[12pt,a4paper]{article}
\usepackage{float}
\usepackage{pgfplotstable,pgfplots,booktabs,filecontents,pdflscape,
csvsimple,longtable}
\newcolumntype{M}{>{$}c<{$}}
\begin{document}
\begin{filecontents}{result3.dat}
model,-1,-2,-3,-4,-5,-6,-7,-8,-9,-10
fdcexp,0.402**,,,,,0.402**,,,,
,-2.19,,,,,-2.19,,,,
L.gdpper,-0.00472,-0.00465,-0.00540***,-0.00542***,-0.00540***,-0.00472,-0.00465,-0.00540***,-0.00542***,-0.00540***
,(-1.63),(-1.60),(-3.45),(-3.45),(-3.44),(-1.63),(-1.60),(-3.45),(-3.45),(-3.44)
growth,0.138,0.149,0.0726,0.0794,0.0868,0.138,0.149,0.0726,0.0794,0.0868
,-0.29,-0.32,-0.18,-0.2,-0.22,-0.29,-0.32,-0.18,-0.2,-0.22
D.lnhuman,-0.533,-0.496,-0.639,-0.661,-0.622,-0.533,-0.496,-0.639,-0.661,-0.622
,(-0.60),(-0.55),(-0.90),(-0.93),(-0.88),(-0.60),(-0.55),(-0.90),(-0.93),(-0.88)
ligdp,-0.251,-0.238,-0.148,-0.135,-0.148,-0.251,-0.238,-0.148,-0.135,-0.148
,(-1.55),(-1.49),(-0.58),(-0.53),(-0.58),(-1.55),(-1.49),(-0.58),(-0.53),(-0.58)
fdcall,,0.400**,,,,,0.400**,,,
,,-2.21,,,,,-2.21,,,
fdcave,,,0.0626*,,,,,0.0626*,,
,,,-2.06,,,,,-2.06,,
fdcnet,,,,0.0262**,,,,,0.0262**,
,,,,-2.2,,,,,-2.2,
fdcrev,,,,,0.0643***,,,,,0.0643***
,,,,,-3,,,,,-3
yr2007,,,,,,3.495***,3.495***,3.416***,3.398***,3.430***
,,,,,,-9.26,-9.27,-6.02,-5.97,-5.98
_cons,8.796***,8.526***,7.695**,7.575**,7.678**,8.796***,8.526***,7.695**,7.575**,7.678**
,-4.23,-4.17,-2.25,-2.21,-2.25,-4.23,-4.17,-2.25,-2.21,-2.25
N,259,259,259,259,259,259,259,259,259,259
r2_o,0.152,0.131,0.187,0.181,0.185,0.152,0.131,0.187,0.181,0.185
\end{filecontents}
\begin{landscape}
\centering\small
\setlength\tabcolsep{2.2pt}
\pgfplotstabletypeset[
col sep = comma,
display columns/0/.style={string type,,string replace*={_}{\_},column type={l}},
columns/modelx/.style={string type},
]{result3.dat}
\end{landscape}
\end{document}

I would like my table to look like this

DB225681
  • 147
  • 1
    what does "won't work anymore" mean? do you get a TeX error? if so show the exact error that you get, also preferably show an example document that allows the error to be reproduced so it can be debugged. – David Carlisle May 09 '17 at 18:38
  • See https://tex.stackexchange.com/questions/321346/how-to-read-a-variable-from-a-file-in-latex perhaps? The answers employ packages like readaray and datatool. – Steven B. Segletes May 09 '17 at 18:42
  • @DavidCarlisle: I added the code that I used with the data. – DB225681 May 09 '17 at 18:50
  • Do you need merely to "import" the data (subsequently accessible by row,column entry), or do you need to process and render the complete array in tabular form? – Steven B. Segletes May 09 '17 at 18:57
  • the posted code doesn't give the error in your question but ! Package PGF Math Error: Could not parse input '0.402**' as a floating point number, sorry. The unreadable part was near '**'.. isn't this the same error message as your last question, 0.402** wouldn't be a valid floating point number in most languages – David Carlisle May 09 '17 at 19:01
  • If I remove the * I get ! Package PGF Math Error: Could not parse input '(-1.63)' as a floating point number, sorry. The unreadable part was near '(-1.63)'.. so if I remove the () I get ! Missing $ inserted. so if I chnage _ to \_ then your document runs without error so there is nothing to suggest it is a problem with large data sets just that your entries are not simple numbers so you need to clean up or filter the data before plotting – David Carlisle May 09 '17 at 19:06
  • do you just want to make a table or are you going to need to plot the data (if just typesetting a table you don't need pgfplotstable) – David Carlisle May 09 '17 at 19:09
  • @DavidCarlisle I uploaded the format of the table I would like to make from these data. As you can see, the table would have *** and other characters, even a math formula. Do you have any idea I could add these character into the table? – DB225681 May 09 '17 at 19:13

1 Answers1

2

A couple of minutes editing makes something that can be typeset as a table. Your use of the pgfplotstable package implies tex needs to treat the data as numeric values in which case more filtering would be necessary.

here I just changed , to & and added \\ at ends of lines,

It is still over full but before shrinking the fonts and other things do you need so many decimal places, values like -0.00540 make it hard to show reasonable table. You could format these numbers in tex but probably simpler to do that in whatever application is outputting this data.

enter image description here

\documentclass[12pt,a4paper]{article}

\usepackage{pdflscape,longtable,array,dcolumn}
\newcolumntype{M}{>{$}c<{$}}
\begin{document}
\begin{landscape}\small
\setlength\tabcolsep{4pt}
\setlength\LTleft{-10pt}
\setlength\LTright{-10pt plus 1fill}
\begin{longtable}{@{}l*{10}{D..{3.6}}@{}}
model&-1&-2&-3&-4&-5&-6&-7&-8&-9&-10\\
fdcexp&0.402**&&&&&0.402**&&&&\\
&-2.19&&&&&-2.19&&&&\\
L.gdpper&-0.00472&-0.00465&-0.00540***&-0.00542***&-0.00540***&-0.00472&-0.00465&-0.00540***&-0.00542***&-0.00540***\\
&(-1.63)&(-1.60)&(-3.45)&(-3.45)&(-3.44)&(-1.63)&(-1.60)&(-3.45)&(-3.45)&(-3.44)\\
growth&0.138&0.149&0.0726&0.0794&0.0868&0.138&0.149&0.0726&0.0794&0.0868\\
&-0.29&-0.32&-0.18&-0.2&-0.22&-0.29&-0.32&-0.18&-0.2&-0.22\\
D.lnhuman&-0.533&-0.496&-0.639&-0.661&-0.622&-0.533&-0.496&-0.639&-0.661&-0.622\\
&(-0.60)&(-0.55)&(-0.90)&(-0.93)&(-0.88)&(-0.60)&(-0.55)&(-0.90)&(-0.93)&(-0.88)\\
ligdp&-0.251&-0.238&-0.148&-0.135&-0.148&-0.251&-0.238&-0.148&-0.135&-0.148\\
&(-1.55)&(-1.49)&(-0.58)&(-0.53)&(-0.58)&(-1.55)&(-1.49)&(-0.58)&(-0.53)&(-0.58)\\
fdcall&&0.400**&&&&&0.400**&&&\\
&&-2.21&&&&&-2.21&&&\\
fdcave&&&0.0626*&&&&&0.0626*&&\\
&&&-2.06&&&&&-2.06&&\\
fdcnet&&&&0.0262**&&&&&0.0262**&\\
&&&&-2.2&&&&&-2.2&\\
fdcrev&&&&&0.0643***&&&&&0.0643***\\
&&&&&-3&&&&&-3\\
yr2007&&&&&&3.495***&3.495***&3.416***&3.398***&3.430***\\
&&&&&&-9.26&-9.27&-6.02&-5.97&-5.98\\
\_cons&8.796***&8.526***&7.695**&7.575**&7.678**&8.796***&8.526***&7.695**&7.575**&7.678**\\
&-4.23&-4.17&-2.25&-2.21&-2.25&-4.23&-4.17&-2.25&-2.21&-2.25\\
N&259&259&259&259&259&259&259&259&259&259\\
r2\_o&0.152&0.131&0.187&0.181&0.185&0.152&0.131&0.187&0.181&0.185\\
\end{longtable}
\end{landscape}
\end{document}
David Carlisle
  • 757,742