4
$\begingroup$

I need to estimate a multivariate VECM-GARCH (or simply VAR-GARCH) in R.

Browsing on the internet, I did not find anything yet.

Do you know if such kind of packages exists?

Please, note that a BEKK approach is required since I am working on some optimal hedge ratio calculation and volatility analysis.

Any hint will be appreciated.

$\endgroup$
3
$\begingroup$

Yes, it exists and it is called ccgarch package.

You can install that by simply running in R install.packages("ccgarch") and learn more about that on the CRAN relative paper.

Moreover, I suggest you to read this lecture hold by the author during an R conference.

Hope this help.

| improve this answer | |
$\endgroup$
  • $\begingroup$ Thanks for the advice. I went through the package and its slides. it is mentioned that the covariance estimation has direct vecm and bekk representation, yet the result seems not to be a vecm-garch (bekk). Do you know if this package is following the BEKK approach? $\endgroup$ – Hamed Jul 7 '15 at 8:47
  • $\begingroup$ This is the answer from ccgarch author: The ccgarch package is designed only for estimating constant and dynamic conditional correlation GARCH models, so that it is not able to estimate BEKK or vech GARCH model. $\endgroup$ – Hamed Jul 18 '15 at 5:44
0
$\begingroup$

I tried using mgarchBEKK (or mgarch) but it seems like the package firstly estimate the VECM model, then use the residuals (Epsilon t) of the VECM (and their variances) in estimating the BEKK-GARCH model. I believe the correct method is to run the two models as a system, but I do not know how to proceed! Can anyone give me a hint please?

| improve this answer | |
$\endgroup$
0
$\begingroup$

The MTS package has the fucntion to fit a BEKK(1,1) but i would recommend that you use a dcc-garch which has less problems that the BEKK approach an it is faster to run

| improve this answer | |
$\endgroup$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.