At 11:20 AM 1/1/2004 -0500, Michel Camus wrote:
I experience the same "problem" as you do, Paul.
I believe it has to do with "binary logic". The logical condition
("chol") always evaluates to 0 or 1, and nothing else, it never
evaluates to "9" even if the original variable does have "9" as one
value. The same happens even if your missing values in "chol" were
already coded as "." See
http://www.stata.com/support/faqs/data/values.html by W. Gould on
"Logical expressions and missing values". The way to have missing
values is tocondition your expression as follows :
However, the docs say this about the cond command:
"cond(x,a,b,c) returns a if x evaluates to true (not 0), b if x
evaluates to false (0), and c if x evaluates to missing (.)."
So, it seems it should work (assuming his md value is ., or .a, or .b,
etc.; if he is trying to use 9 as an md value, that won't work, he has
to be using one of Stata's md values). Plus, it did work for me! So,
my guess is that either (a) he is using some earlier version of Stata
which does not support this variation of the cond command, or (b) he's
not using one of Stata's md values, e.g. he is using 9 for md when
Stata wants ., .a, etc.
In any event, it only takes one additional line of code to fix, but it
seems like the command as originally written should work except under
the conditions I have just mentioned.
-------------------------------------------
Richard Williams, Associate Professor
OFFICE: (574)631-6668, (574)631-6463
FAX: (574)288-4373
HOME: (574)289-5227
EMAIL: [email protected]
WWW (personal): http://www.nd.edu/~rwilliam
WWW (department): http://www.nd.edu/~soc
*
* 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/
--
