Tumer ([email protected]) asked
>I have n equations:
> y(1)=alpha(1)+beta(1)*x(1)
> y(2)=alpha(2)+beta(2)*x(2)
> .
> .
> .
> y(n)=alpha(n)+beta(n)*x(n)
> The theory says that beta(1)+beta(2)+...+beta(n)=0 should hold. So
>I need to estimate these equations together imposing the above
>constraint. According to help the files I must be able to do this
>with "sureg" and "constraint" commands. Imposing restrictions on
>constants seems fine but when I impose any restriction, like the
>one mentioned above, on any set of parameters other then the
>constants I get the message "redundant or inconsistent constraint".
>What can I be doing wrong? Is there any other way of doing this?
>Thank you all beforehand for any help.
Consider the model:
price = alpha1 + beta1*mpg
displacement = alpha2 + beta2*rep78
weight = alpha3 + beta3*turn
with the constraint beta1+beta2+beta3=0
Using the auto dataset we can use -sureg- and -constraint- to estimate the
model:
. clear
. sysuse auto
(1978 Automobile Data)
. constraint define 1 [price]mpg+[displacement]rep78+[weight]turn=0
. sureg (price mpg) (displacement rep78) (weight turn), c(1)
Seemingly unrelated regression
Constraints:
( 1) [price]mpg + [displacement]rep78 + [weight]turn = 0
----------------------------------------------------------------------
Equation Obs Parms RMSE "R-sq" chi2 P
----------------------------------------------------------------------
price 69 1 2666.103 0.1497 37.59 0.0000
displacement 69 1 84.46202 0.1657 4.78 0.0288
weight 69 1 405.2265 0.7349 189.30 0.0000
----------------------------------------------------------------------
----------------------------------------------------------------------------
--
| Coef. Std. Err. z P>|z| [95% Conf.
Interval]
-------------+--------------------------------------------------------------
--
price |
mpg | -106.5862 17.38525 -6.13
0.000 -140.6607 -72.51175
_cons | 8415.248 482.664 17.44 0.000 7469.244
9361.252
-------------+--------------------------------------------------------------
--
displacement |
rep78 | -32.82048 15.01316 -2.19
0.029 -62.24573 -3.395232
_cons | 309.7799 53.54245 5.79 0.000 204.8386
414.7212
-------------+--------------------------------------------------------------
--
weight |
turn | 139.4067 10.13225 13.76 0.000 119.5478
159.2655
_cons | -2515.953 406.1126 -6.20
0.000 -3311.919 -1719.987
----------------------------------------------------------------------------
--
Sincerely,
-- Gustavo
[email protected]
*
* 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/
