I was using the VAR model and to see the impulse response function I used Cholesky decomposition. data is quarterly from 2003q1 to 2019q2. The first variable is oil price as a supply shock, second output gap by applying Hodrick-Prescott filter, third for the monetary policy-money base, fourth nominal effective exchange rate, and fifth cpi. After the first difference, they are stationary. response of CPI to nominal effective exchange rate shock was negative, which means that there is deflation, but at the same time base money has a positive response to neer shock, which means that the national bank uses expansionary policy. and if there is deflation why use national bank expansionary policy?
some people from this site advised me to use the structural VAR model instead of the ordinary VAR model. unfortunately, I do not have experience working with it and I have a problem constructing the model correctly. can anyone help how to construct a correct matrix? my baseline= (oil, output gap, money base, Neer, CPI)
(a 0 0 0 0)
(a a 0 0 0)
(a a a 0 0 )
(0 a a a 0)
(0 a 0 a a)
I should do something like this, but I don't know if it is correct, so please help me to order variables correctly.
Macroeconomics does not provide sufficiently clear-cut suggestions on the restrictions that you impose on the relationship among such variables. If all the variables become stationary following first differencing, why not try a vector error correction (VEC) model if you want your model to possess some desirable long-run equilibrium properties?
What is the aim of your analysis/project?
Why do you want to use VAR?
Why do you use HP-filter, although it has no economic-theoretical base?
I think you should start think over a good specifiction before going into econometric estimations/tests-
I published the below paper in 2013. I employed a SVAR model with the same variables that you plan to use. Good luck!
Article Evaluating the Effects of Monetary Policy Shocks on GCC Countries
Hi Irakly
you may try changing the order of the variables. In SVAR models with chokesky decomposition the order of vars is such that the first variable is affected only by its own shocks and not CONTEMPORANEOUSLY affected by shocks to the other variables. the second one, is affected only by its own shocks plus the shocks affecting the first one. The third one by the shocks affecting the first two plus is own. So on and so forth... I would revise your ordering. you can see papers by stock and Watson for author active references on this field.
good luck!
Dear Irakhi.
There are laws that guide the use of SVAR.
1. Theory
2. Empirical findings
3 Institutional knowledge
The major difference between SVAR and VAR is the contemporaneous (instantaneous) relationship among the variables.
SVAR can be recursive or non recursive. It all depends on your choice.
Therefore, I recommend the comment and articles @Sayyed Mahdi Ziaei and @ Benedetto Molinari.
Regards
Hello,
You can follow the link below about the SVAR:
https://www.youtube.com/watch?v=hwB2TbhqgH0
and
See Narayan et al.'s (2008) Article A structural VAR analysis of electricity consumption and rea...
I hope it works!
Best.
- Badges
- Science topic
- Similar topics
- Philosophy
- Esthetics
- Interpretation