How can I influence the order of variables in a VAR Model for solving the identification problem in the irf?
7 views (last 30 days)
Show older comments
Hi,
I am analyzing some financial time series with a VAR Model, and I would like to produce an orthogonalized IRF ordering the variable accountable for the IRF shock of relevance last in the vector of endogenous variables.
Is the only thing I have to do to account for this to put the respective variable last in the code when providing the model with my data? Taking as an example the case study from https://de.mathworks.com/help/econ/varm.estimate.html:
EstMdl = estimate(Mdl,[rcpi unrate])
In this case, would unrate be the varibale ordered last in the Vector of endogenous variables in the model, lets call it Y_t? I assume this is the case, but I am not certain about this and would like to make sure I am correct. In the subsection Algorithms and in the case study of https://de.mathworks.com/help/econ/varm.irf.html, there is some reference to variable orders on VARs, but I cannot find explicit info on this.
Thanks a lot for your help; please apologize for any potential lapse in my post.
Best
Answers (0)
See Also
Categories
Find more on Vector Autoregression Models in Help Center and File Exchange
Community Treasure Hunt
Find the treasures in MATLAB Central and discover how the community can help you!
Start Hunting!