Hello everyone,

Hope everything is going well.

I have a question about how to identify the matched firm. The sample data is the following.

I want to identify the target firms (target=1) that match with focal firms (target=0) by the following criteria:


1. target firms (target=1) and focal firms(target=0) are in the same industries (same industry code);

2. the Size different between target firms (target=1) and focal firms(target=0) is less than 5% ( |% diff in Size| < 5% of focal firms (target=0)).


For example, for firms in 2009, target firm (target = 1, symbol = 466) is matched with focal firm (target = 0, symbol = 1889). Because these two firms share same industry code (industry = "C"). Also, target firm size (target = 1, symbol = 466, Size = 10.98468) is within 5% of focal firm size (target = 0, symbol = 1889, Size = 10.68249). In the other word, |10.98468 - 10.68249| < 5%*10.68249.

Much appreciated for your help, in advanced. And have a lovely day.

Cheers,
Frank


Code:
* Example generated by -dataex-. For more info, type help dataex
clear
input double symbol float(year lag_Size target) str1 industry
   353 2009 11.908743 1 "C"
   430 2009  14.15009 1 "C"
   466 2009 10.984676 1 "C"
   469 2009  9.873674 1 "C"
   526 2009 13.150103 1 "C"
   631 2009  12.35215 1 "C"
   709 2009 12.046097 1 "C"
  1889 2009 10.682493 0 "C"
  1981 2009 11.658934 1 "I"
599931 2009  11.79596 1 "D"
600182 2009  11.63502 0 "F"
600289 2009 13.280113 0 "B"
600695 2009 13.557528 1 "S"
   288 2010  14.53094 1 "C"
   365 2010 11.574177 1 "C"
   381 2010 12.807743 0 "C"
   602 2010  10.66294 0 "L"
   652 2010 11.655706 0 "C"
  1817 2010  11.09666 1 "C"
  1998 2010  9.882786 0 "C"
  2017 2010 11.316225 0 "C"
  2063 2010 11.951778 0 "C"
  2185 2010 14.827463 1 "C"
299851 2010  11.65016 0 "I"
299886 2010 11.679768 0 "C"
599925 2010 12.312794 . "G"
599928 2010 12.259146 1 "F"
600302 2010  12.15644 1 "E"
600488 2010 14.607515 1 "C"
600771 2010 13.569136 1 "B"
601527 2010 15.686802 1 "C"
   201 2011  14.60073 0 "C"
   313 2011 12.489162 1 "C"
   431 2011 13.445896 0 "K"
   606 2011 12.256338 0 "C"
   635 2011 10.814608 0 "C"
  1916 2011  11.85497 1 "M"
  1919 2011 11.543577 1 "C"
  1951 2011 11.106693 0 "C"
  2117 2011 11.584675 1 "C"
  2235 2011 11.244164 0 "C"
  2301 2011 11.771706 0 "C"
299827 2011 11.895157 0 "R"
299843 2011  10.90149 0 "C"
299916 2011 10.531273 0 "C"
299927 2011 11.159685 1 "C"
299943 2011 11.030184 0 "A"
299945 2011 11.296352 0 "C"
300000 2011  12.55377 1 "C"
599855 2011  11.19978 0 "C"
600537 2011 13.267574 1 "C"
600601 2011  14.07104 1 "C"
600767 2011 12.057472 1 "C"
   344 2012  11.50715 1 "D"
   535 2012  12.47966 0 "A"
   556 2012 12.291164 1 "C"
  1905 2012   11.5334 1 "C"
  2109 2012  13.20733 0 "C"
  2121 2012 12.041492 0 "A"
  2299 2012 11.014174 0 "C"
  2330 2012 10.996252 0 "C"
  2412 2012 12.061598 1 "C"
300120 2012  12.92989 1 "C"
300120 2012  12.92989 1 "C"
599920 2012 13.254177 1 "F"
599992 2012 12.251202 1 "C"
600056 2012 14.127745 0 "S"
600333 2012  13.56007 1 "K"
600350 2012 14.038223 1 "C"
600475 2012 14.364258 1 "K"
600479 2012 11.347134 1 "C"
600664 2012  12.63239 1 "D"
600807 2012 12.063375 1 "H"
601315 2012 14.573185 1 "C"
601416 2012 12.484383 0 "C"
601677 2012  13.31427 1 "C"
   321 2013 13.256262 0 "C"
   483 2013  13.49034 1 "C"
   766 2013 13.705702 1 "D"
  1848 2013  12.84603 0 "C"
  1872 2013 11.593857 0 "C"
  1950 2013 11.065273 1 "C"
  1970 2013 11.810514 0 "C"
  2082 2013 11.373936 0 "C"
  2383 2013 12.126085 0 "C"
  2411 2013  11.05875 0 "C"
299935 2013  11.59205 1 "C"
299981 2013  11.04519 0 "C"
599943 2013 13.732157 0 "C"
600091 2013 13.992184 0 "C"
600298 2013  13.71964 1 "C"
600310 2013 13.627623 0 "K"
600317 2013 11.963255 1 "C"
600485 2013 13.576433 1 "C"
600502 2013 12.433384 0 "C"
600519 2013  11.57269 1 "D"
600546 2013  10.85954 1 "C"
600680 2013  12.33401 1 "R"
600682 2013  11.80856 0 "B"
600770 2013 14.250835 0 "C"
end
------------------ copy up to and including the previous line ------------------