Many thanks - What I wanted to do was set up significance limits for a
parameter (as opposed to confidence limits), for which I'd need to force
the regression coeffcient to be whatever it should be under H0
(typically 0). I'd looked at help constraint but didn't find it that
helpful; the example I tried was with regress (which doesn't support
constraints) so I'll try again with glm. Using the test option might
work if I can extract the F value (though I'd need the se|H0 as well)
Paul Silcocks BM BCh, MSc, FRCPath, FFPH, CStat Clinical Senior Lecturer
Nottingham Clinical Trials Support Unit
Room B39
School of Community Health Sciences
University of Nottingham Medical School
Nottingham
NG7 2UH
tel +44 115 9515151 x 30514
Direct line: +44 115 8230505.
General office: +44 115 8230500
fax +44 115 8230515
e-mail paul.silcocks@nottingham.ac.uk
-----Original Message-----
From: owner-statalist@hsphsun2.harvard.edu
[mailto:owner-statalist@hsphsun2.harvard.edu] On Behalf Of Richard
Williams
Sent: 11 May 2007 21:39
To: statalist@hsphsun2.harvard.edu
Subject: Re: st: regression with coefficients at a set value
At 07:07 AM 5/11/2007, Silcocks Paul wrote:
>What is the best way to perform a multiple regression-type analysis
>(Cox, Poisson, glm etc) with some or all of the regression coefficients
>constrained to a fixed value (not necessarily all the same)?
I'm not clear on the question. Do you mean, how do you impose
constraints? If so, see
help constraint
and also check the help for whatever command you are using. Not all
commands support a constraints option, but sometimes there is a similar
command that will. For example, ologit does not support the constraints
option but the user-written -oglm- does (rather slowly, mind you, but it
does support them). Also, Jeroen Weesie's -linest- command can be quite
handy for commands that do not otherwise support constraints.
Also, depending on your purposes, some commands support an offset and/or
exposure option.
Also, the -test- command includes a -coef- option, which reports
estimated constrained coefficients, e.g.
. quietly reg price mpg weight
. test weight=2, coef
( 1) weight = 2
F( 1, 71) = 0.16
Prob > F = 0.6939
Constrained coefficients
------------------------------------------------------------------------
------
| Coef. Std. Err. z P>|z| [95% Conf.
Interval]
-------------+----------------------------------------------------------
-------------+------
mpg | -22.03124 50.85898 -0.43 0.665 -121.713
77.65053
weight | 2 . . . .
.
_cons | 595.5436 1121.893 0.53 0.596 -1603.325
2794.413
------------------------------------------------------------------------
------
If none of this answers your question, please clarify what the question
is.
-------------------------------------------
Richard Williams, Notre Dame Dept of Sociology
OFFICE: (574)631-6668, (574)631-6463
HOME: (574)289-5227
EMAIL: Richard.A.Williams.5@ND.Edu
WWW: http://www.nd.edu/~rwilliam
*
* 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/
This message has been checked for viruses but the contents of an attachment
may still contain software viruses, which could damage your computer system:
you are advised to perform your own checks. Email communications with the
University of Nottingham may be monitored as permitted by UK legislation.
*
* 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/