Bill Rising
> I've run into a situation where I'd like to build an
> expression which
> contains different variables depending on the observation.
> Now, this
> could be done by nesting a series of -cond()-s, though it
> would be a bit
> complex (and hard to generalize).
>
> Is there some way to use variables a pointers (use indirect
> evaluation of
> the variables)? For example suppose that there were three variables:
>
> bleen bling blam
> ing 1 11
> am 2 12
> ing 3 13
> ing 4 14
>
> I'd like to be able to have
>
> gen blob = eval("bl"+bleen), so that the dataset would then be
>
> bleen bling blam blob
> ing 1 11 1
> am 2 12 12
> ing 3 13 3
> ing 4 14 4
>
> where the (non-existant) -eval()- function takes its string
> argument and
> interprets it as the name of a variable.
>
> I vaguely recall being stumped by this once in the past,
> and know that
> I'm stumped once again. Any tips would be much appreciated,
> even negative
> ones.
As Bill says, there is a -cond()- solution:
gen blob = cond(bleen == "ing", bling, blam)
Macros could also be used with the downside
of a great loss in speed:
gen blob = .
forval i = 1 / `= _N' {
replace blob = bl`= `bleen'[`i']' in `i'
}
Nick
[email protected]
*
* For searches and help try:
* http://www.stata.com/support/faqs/res/findit.html
* http://www.stata.com/support/statalist/faq
* http://www.ats.ucla.edu/stat/stata/
