Thanks to Richard for warning us about this little quirk re -lincom- after
-svy- commands in Stata 8. This is controlled by the undocumented Stata 8
command -is_svy-, which you can find out about by typing
whelp is_svy
I think I will do something about the -svy- feature in the coming Stata 8
version of -lincomest-, which I will write after fixing the feature re
-logistic- in the final Stata 7 version of -lincomest-. In the meantime, I
am scanning through the Stata 8 -lincom- and seeing if I can find any more
hazardous features like these.
Roger
At 20:36 06/05/2005, Richard Williams wrote:
At 08:20 PM 5/6/2005 +0100, Roger Newson wrote:
Hello All (especially StataCorp)
I am still using Stata 8.2 under Windows 2000, and have just found out
about the behaviour of -lincom- if invoked after -logistic-, whereby it
returns in -r()- an exponentiated estimate and its asymptotic standard
error, even if the user has not specified any of the options -eform-,
-or-, -hr-, -irr- or -rrr-. (See -[R] lincom- in the Stata 8 manuals.)
This is a problem, because my own -lincomest- package (downloadable from
SSC) invokes -lincom- without any estimation options, and attempts to use
the estimate -r(estimate)- and its standard error -r(stderr)- to
calculate a 1x1 matrix of estimates and a 1x1 variance matrix to be
stored in the estimation results, and then lists the estimation results,
using -eform()- only if specified by the user. (See Newson (2003) for
more about -lincomest-.)
Further, perhaps I am missing it, but I don't see anything like an xb
option that could force the behavior you want. I suppose you would have
to add lines to your program to make sure it was doing what you wanted.
Another little quirk I discovered in Stata 8: After most commands, lincom
returns its result in r(estimate). But, after svy commands, it returns
its result in r(est). This was screwing up one of my programs until I
figured out what was going on. I asked tech support if there was any
particular reason for this and they said no, but they added that this
would not be an issue in Stata 9.
I think with returns, consistency is very important. With regular
commands, you can often support both old and new syntax; but if you change
the behavior/names of returns or do things inconsistently across programs
it can create problems.
-------------------------------------------
Richard Williams, Notre Dame Dept of Sociology
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/
--
Roger Newson
Lecturer in Medical Statistics
Department of Public Health Sciences
Division of Asthma, Allergy and Lung Biology
King's College London
5th Floor, Capital House
42 Weston Street
London SE1 3QD
United Kingdom
Tel: 020 7848 6648 International +44 20 7848 6648
Fax: 020 7848 6620 International +44 20 7848 6620
or 020 7848 6605 International +44 20 7848 6605
Email: [email protected]
Website: http://phs.kcl.ac.uk/rogernewson/
Opinions expressed are those of the author, not the institution.
*
* 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/
