Timothy Purcell,
I do not have an answer to your question. But would like to see a summary of the responses.
I have a query. You wrore:
I can also run a VEC, but
the variable of interest is Y1 and this will only be included as an
exogenous or deterministic variable and I won't be able to calculate the
impulse response function of this variable.
I do not follow this point. If I remember correctly, I(0) variables can also be added to a VEC
and their effect would be to increase the cointegration rank by 1 for every such variable.
So in a cointegrated VAR model, you also get Y1 as endogenous, is it not?
Regards,
-- Krishnadas
------------------------------------------------------------------------------------------------------------------
C. K. Krishnadas [log in to unmask] c k krish at cyberspace dot o r g
http://www.cyberspace.org/~ckkrish Alt. email: [log in to unmask]
------------------------------------------------------------------------------------------------------------------
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
|