I have four variables in a VEC model. I have allready established the presence of three vectors.
My hypothesis is that variable 3->variable 1, variable 3-> variable 2, and that variable 4-> variable 3. The big question now is, how do I set all this up with restrictions? I use Eviews (and Gretl).
I am really thankful for productive advice on this!
Best regards,
Lars
Restrictions may be placed on the coefficients B(r,k) of the r-th
cointegrating relation:
B(r,1)*variable 1 + B(r,2)*variable 2 +
B(r,3)*variable 3+ B(r,4)*variable 4
Restrictions may also be put on the adjustment coefficients, where
A(k,r) is the coefficient of the r-th cointegrating relation in the k-th
VEC equation, and where:
k = 1 D(variable 1) equation
k = 2 D(variable 2) equation
k = 3 D(variable 3) equation
k = 4 D(variable 4) equation
This is in relevance to Eviews
Thank you. How do I write the restrictions and where do I put them (0,1,-1 etc)?
If you are working on Eviews the advise of Mousumi is correct. He is saying exacty how to put the restriction in Eviews.
How many cointegration relations do you have? The columns of variables coefficient are first tested for stationarity using ADF. Only the stationary columns form cointegrating vectors. The coefficients of cointegrating vectors are written in linear equations and absolute values of the coefficients are presented in a matrix form and a chi-square test is done test the significance.
- Badges
- Science topic
- Similar topics
- Vectorization