You can use a macro extended function to set a macro equal to the type
of each variable in turn, and drop the ones whose type includes "str"
(to drop string variables):
drop _all
set obs 10
gen x=1
gen y="a"
gen str8 z="b"
foreach v of varlist _all{
local t:type `v'
di "`t'"
if index("`t'", "str")~=0{
drop `v'
}
}
Jamie Griffin
>>> [email protected] 09/14/05 6:36 pm >>>
I have a quick question about how to select variables by their format.
Suppose I want to drop (or rename, or do something else) with all
string variables. Clearly,
-drop if- doesn't work as that selects observations, not variables.
The motivation for this question is that I want to collapse a whole
dataset by some variable:
collapse *, by(somevar)
Specifically, I want to get the mean of all numeric variables by
somevar, and get the highest "value" in alphabetical order of the
string variables by somevar. So if there is no way to do this in one go
using -collapse-, I would collapse separately for the numeric and the
string vars, and for that purpose I need to find a way to select
variables by their format.
Thanks for any guidance on this!
Tewodaj
*
* 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/
*
* 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/
