[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]
st: imposing a constraint on an endogenous variable when using irf createin Stata 9
I am trying to find a way to impose a constraint on an endogenous
variable when Stata computes impulse response functions using irf create.
I estimated a SVAR of 5 variables, and would like to find a way to force
one of the variables to remain constant when the impulse response
function is computed. This is following Sims and Zha (1996).
I tried manipulating the SVAR results: I modified e(b_var) and e(b) such
that zeros replaced all the coefficients (respectively, lagged and
contemporaneous) of the variable in question, using the ereturn command.
I then created another impulse response function using irf create.
I succeeded in suppressing the variable's response to lagged values -
the modification of e(b_var) was effective - but the contemporaneous
response remained the same - the modification of e(b) did not have any
This method is crude, yet I couldn't find a better way - I tried the
statalist archives, Googling, as well as the manuals. This method also
does not allow me to impose a constraint for some of the steps in the
irf, but not for others.
If anyone has a piece of advice for me, I would be very grateful.
* For searches and help try: