Plotting

[Yun Dong Yeo]

Dear Statalist users

I have a simple question.
The current data I have looks something like the below.
The data involves firms (panelid) engaging in international collaboration. The Nation denotes the home country of the firm while the LV_country denotes the host country (home country of partner firm).

I'm not sure if this is something possible, but I would like to examine (if possible, plot) the Nation-LV_country for all my observations to see if certain cluster or patterns emerge.
I tried exploring network analysis, but it seems network analysis may not be applicable because the data is not in a dyad format.
Would there be any ways to do this?

Any advice, input, or help would be greatly appreciated.
Thank you.
Sincerely, Yun

[Nick Cox]

The number of distinct pairs is the maximum given by

egen group = group(Nation LV_country) 
su group

and that number -- and the number of distinct countries in either variable -- may rule most useful plots.

[Yun Dong Yeo]

Thank you very much for your response Nick!