Help ARDL model and coefficients of long run relationship
Posted: Mon Apr 02, 2012 11:36 pm
Hi everybody, I'm a student and I'm working with an ARDL model to estimate the coefficient of the long-run relationship and I have a problem.
Example:
MY model is:
dY= c + c1*Y(-1) + c2*X(-1) + dY(-1) + dX + dX(-1) (*)
(s.e)
t-stat=0.5 for example
I have estimated this model ARDL (2,2) and found out the coefficients. Then I have long run coefficients described as:
Y= c + (-c2/c1)*X
The problem is I don't know which t-stat I have to use to test H0: coefficent of X = 0. I think t-stat will be (-c2/c1)/s.e= 3 for example ( s.e is the value which is estimated from (*)) .
So which one is true ? Please help me, I'm sorry for my bad english
Example:
MY model is:
dY= c + c1*Y(-1) + c2*X(-1) + dY(-1) + dX + dX(-1) (*)
(s.e)
t-stat=0.5 for example
I have estimated this model ARDL (2,2) and found out the coefficients. Then I have long run coefficients described as:
Y= c + (-c2/c1)*X
The problem is I don't know which t-stat I have to use to test H0: coefficent of X = 0. I think t-stat will be (-c2/c1)/s.e= 3 for example ( s.e is the value which is estimated from (*)) .
So which one is true ? Please help me, I'm sorry for my bad english