trouble with code

Good day,

I have a portion of a code below. I have estimated a dynamic model.

yt = A1yt(-1) + A2xt(1) + A3xt(2) + A4xt(3)

I am trying to retrieve the long-run estimates of the parameters and their standard errors

a2 = A2/(1-A1); a3 = A3/(1-A1); a4 = A4/(1-A1)

I believe this is what the code below is trying to do; But for some reason, the long run estimates (ai's) appear to be the same as the short run's (Ai'). I have been trying to figure out why but can't appear to find where the error is. Any help? please?

Thanks,

Ozey

/* Long-run parameter estimates and s.e.";*/

/* for group1*/
bt1 = ththat1[3:k2+1]; phi1=ththat1[1];

p1 = (eye(k2-1)~(bt1./(1-phi1)))./(1-phi1);

bse1=p1*sb1[2:k2+1,2:k2+1]*p1';

/* for group2*/
bt2 = tht2hat[2:rows(tht2hat)]; phi2=tht2hat[2];

p2 = (eye(k2-1)~(bt2./(1-phi2)))./(1-phi2);

bse2=p2*se2*p2';

bt = bt1|bt2; 
bse = sqrt(diag(bse1))|sqrt(diag(bse2));
bt'|bse';

1 Answer



0



You are much more likely to get a helpful answer if you tell people specifically: which variables in the code correspond to which variables in the equations, what you expect from different sections of the code and what you actually found.

aptech

1,773

Your Answer

1 Answer

0

You are much more likely to get a helpful answer if you tell people specifically: which variables in the code correspond to which variables in the equations, what you expect from different sections of the code and what you actually found.


You must login to post answers.

Have a Specific Question?

Get a real answer from a real person

Need Support?

Get help from our friendly experts.

Try GAUSS for 14 days for FREE

See what GAUSS can do for your data

© Aptech Systems, Inc. All rights reserved.

Privacy Policy