----- Original Message -----
From: "John D. Levendis" <[email protected]>
To: "statalist" <[email protected]>
Sent: Thursday, June 26, 2003 7:02 PM
Subject: st: constrained VAR
> Greetings statalisters,
>
> I've got a question regarding constraints on variables in a VAR.
>
> Consider the VARX below:
>
> P = B1*P(t-1) + B2*Q(t-1) + B3*X1(t) + B4*X2(t) + e(t)
> Q = B5*P(t-1) + B6*Q(t-1) + B7*X1(t) + B8*X2(t) + e(t)
>
> It can be estimated by
> -var P Q, exog(X1 X2) lags(1)
>
> I want to constrain the coeff on X1 in the Q equation to be zero.
>
> If the Xs were endogenous variables, I could have done:
> constraint define 1 [Q] X1 = 0
> var P Q, exog(X1 X2) constraint(1)
>
> This doesn't seem to work. I get the error message
> "regressor Q not found"
> "at least one constraint in constraints(1) invalid"
>
> Is there a way to constrain the X1 if its an exogenous variable?
>
> Thanks in advance.
>
> --John
John,
I'm not sure why your procedure is not working , but that is how you would
contrain the variable in -var-.
Scott
Example
. use http://www.stata-press.com/data/r8/lutkepohl.dta
(Quarterly SA West German macro data, Bil DM, from Lutkepohl 1993 Table E.1)
. gen x1 = invnorm(uniform())
. gen x2 = invnorm(uniform())
. constraint 1 [dlincome]x1 = 0
. var dlinv dlinc , exog(x1 x2) constraint(1)
Estimating VAR coefficients
Iteration 1: tolerance = .0037264
Iteration 2: tolerance = .0000714
Iteration 3: tolerance = 1.373e-06
Iteration 4: tolerance = 2.641e-08
Vector autoregression
Constraints:
( 1) [dlincome]x1 = 0
Sample: 1960q4 1982q4
--------------------------------------------------------------------------
Equation Obs Parms RMSE R-sq chi2 P
--------------------------------------------------------------------------
dlinvestment 89 7 .042516 0.1052 10.16803 0.1178
dlincome 89 6 .010981 0.1184 11.95642 0.0354
--------------------------------------------------------------------------
Model lag order selection statistics
------------------------------------
FPE AIC HQIC SBIC LL
Det(Sigma_ml)
2.878e-07 -9.3632914 -9.2055007 -8.9718205 430.66647 2.148e-07
----------------------------------------------------------------------------
--
| Coef. Std. Err. z P>|z| [95% Conf.
Interval]
-------------+--------------------------------------------------------------
--
dlinvestment |
dlinvestment |
L1 | -.2271166 .1065471 -2.13
0.033 -.4359451 -.0182881
L2 | -.1408893 .1064105 -1.32 0.185 -.3494499
.0676714
dlincome |
L1 | .6925616 .3935558 1.76 0.078 -.0787936
1.463917
L2 | .4537439 .3922438 1.16 0.247 -.3150398
1.222528
x1 | .0042531 .0042963 0.99 0.322 -.0041675
.0126738
x2 | .0036471 .0047478 0.77 0.442 -.0056584
.0129525
_cons | .0008888 .0111707 0.08 0.937 -.0210053
.0227829
-------------+--------------------------------------------------------------
--
dlincome |
dlinvestment |
L1 | .0542028 .0272641 1.99 0.047 .0007661
.1076395
L2 | .0521213 .0272318 1.91 0.056 -.0012521
.1054946
dlincome |
L1 | .0296899 .1011143 0.29 0.769 -.1684905
.2278702
L2 | .0714579 .1009812 0.71 0.479 -.1264616
.2693773
x1 | (dropped)
x2 | .0015409 .0012245 1.26 0.208 -.0008591
.0039409
_cons | .015591 .0028791 5.42 0.000 .0099481
.021234
----------------------------------------------------------------------------
--
.
*
* 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/
