Extract IDs where condition is met, looped over multiple time points

[Josephine George]

I have aggregate data summarising individual level data. In the aggregate data, each row describes a combination of country and generation, each variable is a survey round from the orginal data, and the values are the number of responses in a given category and survey round. I have substantially abbreviated the example data here; there are at least 20 survey rounds for some of my data sources.

* Example generated by -dataex-. For more info, type help dataex
clear
input str26 category int(_1997_Oct _1999_March _2000_Nov)
" France prewar"  246 241 243
" France boomers" 408 384 365
" France genx"    323 313 298
" France millen"   28  62  97
" France genz"      0   0   0
end

I am trying to end up with a list of category-survey rounds where there were fewer than n observations. In the above example, if n=100, I would want to end up with something like:

OCt1997	March1999	Nov2000
France millen	France millen	France millen
France genz	France genz	France genz

I do also have the individual level data, currently "long", if what I want can be more easily achieved from that.

[Josephine George]

This turned out to be slightly more straightforward than I thought. In case it helps anyone, this code seems to work:

        ds category, not
        foreach x of varlist `r(varlist)' {
        list category if `x'>0 & `x'<100
        }

Edit: I am still struggling to figure out how to get the output into a more useful format (ideally Excel rather than lists in the Stata window)