Dear Tom:
The BEKK model involves a lot of interaction terms.
How can we use rats to measure the marginal effects of each variable in the BEKK model.
Best regards
Marginal effects in the BEKK model
Re: Marginal effects in the BEKK model
I'm not sure what you mean. You can't easily look at partial derivatives because there aren't good "representative" values for the other components of the model---because everything tends to move together, it's hard to isolate one part.avalokita wrote:Dear Tom:
The BEKK model involves a lot of interaction terms.
How can we use rats to measure the marginal effects of each variable in the BEKK model.
Best regards
Re: Marginal effects in the BEKK model
Dear Tom:
Due to the complexity of BEKK model, we face difficulty in interpreting the coefficients.
Take a simple bivariate BEKK for example.
According to the formula, the variance of the second variable is:
h_(22,t)=c_12^2+c_22^2+b_12^2 h_(11,t-1)+2b_12 b_22 h_(12,t-1)+b_22^2 h_(22,t-1)+a_12^2 e_(1,t-1)^2+2a_12 a_22 e_(1,t-1) e_(2,t-1)+a_22^2 e_(2,t-1)^2
So h_(22,t) is under at least three main effects of the first variable:
(1) variance of the first variable h_(11,t-1): b_12^2 h_(11,t-1),
(2) covariance of these two variables h_(12,t-1): 2b_12 b_22 h_(12,t-1),
(3) error terms of the first variable e_(1,t-1)^2: a_12^2 e_(1,t-1)^2+2a_12 a_22 e_(1,t-1) e_(2,t-1),
Looking at partial derivatives, one can get
(1) 2*b_12
(2) 2b_12 b_22
(3) 2*a_12^2* e_(1,t-1)+2*a_12 a_22 *e_(2,t-1)
Coefficients in traditional regressions tell us the incremental influence of a unit change in one variable, keeping all other variables unchanged.
But how can we interpret the coefficients in BEKK models?
If everything tends to move together, and to isolate one part is difficult, what is the use of coefficients in BEKK?
By the way, the BEKK coefficients [ex. A(1,2), B(2,1)] estimated by RATS mean the influence of squared form ( e_(1,t-1)^2) or raw form ( e_(1,t-1))?
Official User Guide illustrates how to run a model; but we feel puzzled at the result.
Due to the complexity of BEKK model, we face difficulty in interpreting the coefficients.
Take a simple bivariate BEKK for example.
According to the formula, the variance of the second variable is:
h_(22,t)=c_12^2+c_22^2+b_12^2 h_(11,t-1)+2b_12 b_22 h_(12,t-1)+b_22^2 h_(22,t-1)+a_12^2 e_(1,t-1)^2+2a_12 a_22 e_(1,t-1) e_(2,t-1)+a_22^2 e_(2,t-1)^2
So h_(22,t) is under at least three main effects of the first variable:
(1) variance of the first variable h_(11,t-1): b_12^2 h_(11,t-1),
(2) covariance of these two variables h_(12,t-1): 2b_12 b_22 h_(12,t-1),
(3) error terms of the first variable e_(1,t-1)^2: a_12^2 e_(1,t-1)^2+2a_12 a_22 e_(1,t-1) e_(2,t-1),
Looking at partial derivatives, one can get
(1) 2*b_12
(2) 2b_12 b_22
(3) 2*a_12^2* e_(1,t-1)+2*a_12 a_22 *e_(2,t-1)
Coefficients in traditional regressions tell us the incremental influence of a unit change in one variable, keeping all other variables unchanged.
But how can we interpret the coefficients in BEKK models?
If everything tends to move together, and to isolate one part is difficult, what is the use of coefficients in BEKK?
By the way, the BEKK coefficients [ex. A(1,2), B(2,1)] estimated by RATS mean the influence of squared form ( e_(1,t-1)^2) or raw form ( e_(1,t-1))?
Official User Guide illustrates how to run a model; but we feel puzzled at the result.
Re: Marginal effects in the BEKK model
Dear Tom:
I have reviewed relevant topics and found that u mentioned @MVGARCHFOR in replying another's question.
With expertise in Econometrics, You are our savior indeed.
Can we use @MVGARCHFOR to identify the marginal influence of one variable on the other variables in BEKK models?
(Since @MVGARCHFOR converts BEKK to VECH)
Best Regards.
I have reviewed relevant topics and found that u mentioned @MVGARCHFOR in replying another's question.
With expertise in Econometrics, You are our savior indeed.
Can we use @MVGARCHFOR to identify the marginal influence of one variable on the other variables in BEKK models?
(Since @MVGARCHFOR converts BEKK to VECH)
Best Regards.