How about if you regressed expenditures on the socio-demographic variables
and then used -adjust- to get the adjusted means.
For example, using the auto dataset deriving the average weight adjusted for
length and car origin.
. sysuse auto
(1978 Automobile Data)
. qui reg weight length foreign
. adjust length if fore == 0, by(fore) se ci
----------------------------------------------------------------------------
Dependent variable: weight Command: regress
Covariate set to mean: length = 196.13461
----------------------------------------------------------------------------
----------------------------------------------------------
Car type | xb stdp lb ub
----------+-----------------------------------------------
Domestic | 3317.12 (34.7036) [3247.92 3386.31]
----------------------------------------------------------
Key: xb = Linear Prediction
stdp = Standard Error
[lb , ub] = [95% Confidence Interval]
. adjust length if fore == 1, by(fore) se ci
----------------------------------------------------------------------------
Dependent variable: weight Command: regress
Covariate set to mean: length = 168.54546
----------------------------------------------------------------------------
----------------------------------------------------------
Car type | xb stdp lb ub
----------+-----------------------------------------------
Foreign | 2315.91 (53.3538) [2209.52 2422.29]
----------------------------------------------------------
Key: xb = Linear Prediction
stdp = Standard Error
[lb , ub] = [95% Confidence Interval]
.
Hope this helps,
Scott
-----Original Message-----
From: [email protected]
[mailto:[email protected]] On Behalf Of Puneet K Singhal
Sent: Monday, May 03, 2004 10:20 AM
To: [email protected]
Subject: st: Adjusted mean.
For a group of individuals, I have their socio-demographic variables and
health care expenditures per month for a year. I wanted to get their mean
expenditures per month adjusted for sociodemographic variables. How do I do
it? Any help? Thanks.
Puneet
*
* 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/
