Drop variables based on name

[Joan Marti]

Hi, I have a dataset with 100 variables, 50 of which are called Xi_ante and 50 of which are called Xi_post.

I want to delete those with "_post" from my dataset.

I tried using the following command:

global variables [all variables Xi without _ante or _post]

foreach `var' of global variables{
drop `var'_post
}

But I get a syntax error. Is there any way to tell stata to drop the variables that contain "_post" in their name?

Thank you!

[Fei Wang]

drop *_post

[Joan Marti]

easier than I thought... thanks a lot! I was wondering though, how come that command didn't work? I was going to use a similar loop to rename variables

[Fei Wang]

easier than I thought... thanks a lot! I was wondering though, how come that command didn't work? I was going to use a similar loop to rename variables

foreach var of global variables{
    drop `var'_post
}

Actually, you may use -rename group- instead of a loop to rename variable names.

[Carlo Lazzaro]

Joan:
-split- might be another option:

. set obs 100
number of observations (_N) was 0, now 100

. g wanted="Xi_ante" in 1/50

. replace wanted="Xi_post" in 51/100

. split wanted, p(_post)

. drop if wanted1=="Xi"

. drop wanted

. rename wanted1 wanted

.

[Joan Marti]

Very useful suggestions, thank you!!