SVAR - Difference between LR and SR restrictions

[sivakizildag]

Hello,

In order to analyse the interactions between the following data, I made an SVAR:

- tourist arrivals in France --> A_TOT_SA
- accomodation capacity in number of rooms in hotels (in first difference) --> D(OFF_SA)
- a consumer confidence index for OECD countries --> D(E_OECD)

I do not know though whether I should apply long run or short run restrictions.

So my first question is ; what is the difference, in terms of economic interpretation / intuition, between LR and SR restrictions? Which one should be used in which case?


I made two SVAR with the two types of restrictions; you will find the both SVAR enclosed; SVAR_SR restricted by the matrix ZA and ZB and SVAR_LR restricted by ZB only.

For the SVAR_SR, I find the estimated values of the restriction matrixes in the IRF, as expected. Indeed, you will see that the the Response of A_TOT_SA for Shock 2 in the SR_IRF table and the C(3) coefficient SR_SF table are the same; 51294.

This is not the case, for the SVAR_LR. For the same response to the same shock (A_TOT_SA for Shock 2), I see a 5.62E+10 reponse in LR_IRF table. This is the value of matrix B, first row second column (see LR_SF table), but not C(2).

So, what is the use or the purpose of the C(1), C(2) and C(3) coefficients of the restriction matrix ?

And in the LR restriction matrix (matrix ZB), the variables I had restricted to 0 do not "get" a 0 response from the shocked variable. I really do not understand the purpose of the ZB matrix and the C(1 to 3) coefficients then.

Could you please help me, I am really stuck.

Thank you very much in advance!

[sivakizildag]

Anyone to help me?