I have a Stata program that outputs a local scalar of space-separated variable names.
I have to run the program twice on two samples (same dta) and store the union (intersection - variable names appearing in both scalars) as a new space-separated local scalar (for input to another program).
I can't figure out how to split (per spaces) and or test the occurrences of variable names in each.
Stata has a bunch of extended macro functions to use on lists that you can find with help macrolists, where you can see that A & B returns the intersection of A and B. If A="a b c d" and B="b c f g", then A & B = "b c".
This allows you to do something like this:
clear
scalar l1="vara varb varc"
scalar l2="varc vard vare"
local l1 = scalar(l1)
local l2 = scalar(l2)
local inter: list l1 & l2
scalar inter="`inter'"
scalar list inter
You convert the scalars to locals, get their union, and convert that into a scalar. It is probably easier to just modify your code to use locals rather than scalars so you don't have to deal with conversions.