Hello Statalisters,

Firstly, thank you for having a look at my problem - any help or advice you could offer would be greatly appreciated!

I have merged two panel datasets of companies in Sweden, one of all registered companies in Sweden (~5m obs), and another with companies in a startup incubator (35k obs.). Now I would like to find a "twin" company in Sweden for each company from the incubator dataset. That is, I would like to find a non-incubator company to match each of my incubator companies, with perfect matches across four variables. The four variables are: rr01_ntoms, bslov_antanst, br01a_foubautg, br01b_patlic. My objective is to finish with two data files, one with the incubator companies, and another with the "twins".

A colleague suggested that I use -vmatch-, and I have tried with the following code:
Code:
vmatch casevar, g(match) save(match_incubator) f o(rr01_ntoms, bslov_antanst, br01a_foubautg, br01b_patlic 4)
I am not sure I have been using -vmatch- correctly, as the above code ran for four hours generated a file of 55m observations. I have also seen on forums that one could perhaps use merge to achieve this task, but I am uncertain as to how.

I look forward to any help or advice - thank you, again.

Ayrton