RE: st: RE: truncate variable names

["Nick Cox" <n.j.cox@durham.ac.uk>]

You can do exactly that if all the abbreviations to the first 8 letters are distinct. As said, you'd need to check that any way. 

Nick 
n.j.cox@durham.ac.uk 

SCHOUMAKER Bruno

Thanks a lot !

I agree it may seem a bad idea; Actually, I had a program running with 
variables names truncated to 8 characters (exported from SPSS using an 
old stat/transfer version) -  I received a new data set (same data, but 
with additional observations), exported from SPSS with a new version of 
stat/transfer, but with variables names longer than 8. And it would save 
me a lot of time if I could just re run the syntax with exactly the same 
variable names...

Le 26/07/2010 16:43, Nick Cox a écrit :

> renvars, trim(8)
>
> where -renvars- is downloadable from the SJ files. -search renvars- for
> precise location.
>
> Note that the first principles solution
>
> foreach v of var * {
> 	local new = substr("`v'", 1, 8)
> 	rename `v' `new'
> }
>
> is more instructive about principles but will fail part-way through if
> two or more variable names have the same first 8 characters. -renvars-
> warns you of that problem in advance, so that you can rethink strategy.
>
> By the way, this strikes me as mostly a bad idea unless you are
> exporting to some other software that requires it.

SCHOUMAKER Bruno

> I would like to truncate the variable names of my dataset to 8
> characters.
>
> Does anybody know how I can do that in Stata ?

*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   http://www.stata.com/support/statalist/faq
*   http://www.ats.ucla.edu/stat/stata/