Bookmark and Share

Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

st: SVAR estimation with Stata

Subject   st: SVAR estimation with Stata
Date   Tue, 03 May 2011 17:25:20 -0400

Dear all,

I am running Structural VAR model with Stata. Suppose the contemporaneous
coefficient matrix is G0, that is,


where e(t) is the structural error and u(t) is the error of the reduced form

Here is the Stata codes:

matrix A (exactly the G0 matrix)
matrix B

svar `varlist', exog(`dum') lags(1/6) aeq(A) beq(B)
{`dum' refers to the 11 dummy variables treated as exogenous}

Now create the Impulse Response Function:

irf create model2, set(myirf, replace) step(8)
irf table fevd, noci {noci is to suppress confidence bands}
irf graph oirf, impulse(r) response(cpi) {orthogonalized IRF}
irf graph sirf, impulse(r) response(cpi) {structural IRF}

1. Is it the correct way to involve "lags" to the estimation?
2. The convergence is not achieved, any possible reasons for that?
3. When I create the IRF, Stata was stuck and could not move on. Does anybody
know how this might come across?

Thank you very much. Looking forward to any positive reply.


*   For searches and help try:

© Copyright 1996–2017 StataCorp LLC   |   Terms of use   |   Privacy   |   Contact us   |   Site index