Help with -distinct-

[Josef Kannegaard]

I recently installed the -distinct- module and was wondering if there was a way to integrate it into a table to count distinct observations across the values of a second variable (i.e. something to the effect of 'table var1, statistic(distinct var2)'? I know it allows for by, but the variable I'd like to check has over 900 values, so that isn't practical.

[Jared Greathouse]

Can you use the auto dataset to give a small example of what you want?

[Nick Cox]

egen tag = tag(var1 var2)

egen wanted = total(tag), by(var1)

tabdisp var1, c(wanted)

See the 2008 paper behind distinct for more.