The RATS Software Forum

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 @MVGARCH...

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...

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

Dear Tom: When estimating BEKK models, the RATS software produces the results of estimated coefficients in the level form of C(i,j), A(i,j), B(i,j), accompanied with corresponding significance tests. Yet, BEKK is a nonlinear model in which squared parameters matter in its variance equation. Accordin...

Dear Tom: I am still unclear about the interpretation of the BEKK coefficients for they are quite different from those in typical regressions where all the coefficients and variables are in common forms, not squared. Take h11 for example. If we try to express h11 in one single equation, we'll get h1...

Dear Tom: Let's take an example from the other forum, in which codes look like as follows: equation eq1 us #constant us{1} elec{1} d08 equation eq2 elec #constant us{1} elec{1} d08 group ar1 eq1 eq2 garch(p=1,q=1,model=ar1,mv=bek,XREGRESSORS,pmethod=simplex,piters=10,hmatrices=hh,rvectors=rr) / us e...

Dear Tom: I have overcome the problem about "## MAT15. Subscripts Too Large or Non-Positive." Yet, there are still two follow-up questions. First, when dccexamples.rpf use the garch.asc data to run the fourth model (AGDDCC, Asymmetric Generalized Diagonal DCC), the message alwayes appears ...

You're probably not initializing the Q and EE series of matrices so there's no lagged Q and lagged EE at GSTART. I simply use a different data set and didn't modify the essential codes. The initialization works as the same, that is compute gstart=%regstart(),gend=%regend() gset q = qbar gset ee = q...

Dear Tom: I'm tracing the code using a different data set. But when the program runs this line of code compute bfix=sqrt(.85),afix=sqrt(.10) nonlin afix bfix compute gq=%const(0.0) disp gq maximize(start=%(bq=%const(bfix),aq=%const(afix),StartQC()),method=bfgs,title="Two-step DCC") logl gs...

The degrees of freedom on the standard BQ test will be the number of parameters that are there to reduce serial correlation. That would be the sum of the number of AR and MA elements in the mean model which here is 1. Dear Tom: I'm very grateful for your help, but I am still unclear about one point...

Thank you for your instant reply!!! Yeah, I'm reading the Tsay textbook and reviewing the programs provided by Estima. The textbook mentions how to select the best lag orders for the LB test (n =ln T) but doesn't explain too much about the degrees of freedom applied in chi-squared distributions. The...

There's nothing obviously wrong with it, but it's impossible to tell if it's "good" based upon just that limited amount of information. How does it compare with a simpler model like CC: are the added parameters for BEKK worth it? Is asymmetry necessary? Do standardized residuals show any ...