|
|  |
[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]
Re: st: varlist error? or feature?
At 03:33 PM 10/17/2006, Jeph Herrin wrote:
Why have I never noticed this before? Stata 9, updated.
. d com1-com45
storage display value
variable name type format label variable label
-------------------------------------------------------------------------------
com1 byte %8.0g BPH
medcom2 byte %9.0g meds: CA
com2 byte %8.0g CA
medcom3 byte %9.0g meds: CHF/HTN
com3 byte %8.0g CHF/HTN
medcom4 byte %9.0g meds: ESRD
com4 byte %8.0g ESRD
medcom5 byte %9.0g meds: GERD
com5 byte %8.0g GERD
medcom6 byte %9.0g meds: HIV
com6 byte %8.0g HIV
medcom7 byte %9.0g meds: HTN
.
.
.
And I discovered it when
. egen total=rtotal(com1-com45)
totalled all the -medcom- variables as well. How can I reference
just those variables that start with -com-?
thanks,
Jeph
[...]
It should be pointed out that the abbreviation
com1-com45
refers to all variables in the span from com1 to com45, in the set of
variables in their present order (which you can change with -order-
or -aorder-). It has nothing to do with imputing names that fit the
paradigm that you might be supposing. It has always worked that way.
As some other respondents mentioned, what you probably want is com*
-- as long as that covers only the variables you want. If you have
any names such as "common" or "complete" or "compensation", they will
also be picked up by com*. That's why I usually avoid both of these
abbreviation forms. Instead, I use a list that contains the exact
names I want. If is long, I put it in a macro; sometimes it can be
assembled programmatically.
HTH
--David
*
* 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/
