Page 1 of 9

DCCGARCH11

Posted: Tue Mar 04, 2014 1:33 pm
by trubador
This thread is about dccgarch11 add-in.

The add-in allows you to build and estimate Dynamic Conditional Correlation models, which are the more flexible and parameterized class of Multivariate GARCH-family. It is written/designed with primarily educational purposes in mind and therefore some limitations are imposed to ease the estimation and maintain the user-friendliness of the GUI as well:

i) Second or higher order specifications are not allowed.
ii) Estimation is carried out in two-step.
iii) At most 5 series are allowed.

Please read the documentation for further instructions.

Re: DCCGARCH11

Posted: Wed Mar 05, 2014 1:31 am
by ecofin
very useful Add-in 8) . thanks a lot.

Re: DCCGARCH11

Posted: Sat Mar 15, 2014 8:30 am
by theodore04
is the documentation available?


trubador wrote:This thread is about dccgarch11 add-in.

The add-in allows you to build and estimate Dynamic Conditional Correlation models, which are the more flexible and parameterized class of Multivariate GARCH-family. It is written/designed with primarily educational purposes in mind and therefore some limitations are imposed to ease the estimation and maintain the user-friendliness of the GUI as well:

i) Second or higher order specifications are not allowed.
ii) Estimation is carried out in two-step.
iii) At most 5 series are allowed.

Please read the documentation for further instructions.

Re: DCCGARCH11

Posted: Sat Mar 15, 2014 1:17 pm
by trubador
Assuming you have already installed the add-in, the documentation is available from the Manage Add-ins menu.

Re: DCCGARCH11

Posted: Thu Mar 20, 2014 7:50 am
by matco
Hi, I am new to eviews. I would need to simultaneously estimate a VEC and a bivariate DCC-GARCH, I have try to use the add-in for dcc-garch but I cannot get it to work, also I would need the VEC output in line with the dcc output. would I have to right the program myself or is there is any other way?

Re: DCCGARCH11

Posted: Wed Apr 02, 2014 6:09 am
by misscats
Dear trubador
Does this need Eviews 8 or higher to go with?

Thanks.

Re: DCCGARCH11

Posted: Wed Apr 02, 2014 6:42 am
by trubador
Yes, you need EViews version 8 or higher.

Re: DCCGARCH11

Posted: Tue Jun 10, 2014 1:17 am
by econworker
trubador wrote:Yes, you need EViews version 8 or higher.


Dear Trabadur, thanks for this add-in, is there any specific reason that only 5 variables are allowed?

Re: DCCGARCH11

Posted: Tue Jun 10, 2014 2:15 am
by trubador
Have you read the documentation that comes with the add-in?

Re: DCCGARCH11

Posted: Tue Jun 10, 2014 6:38 am
by econworker
[quote="trubador"]Have you read the documentation that comes with the add-in?[/quo

Unfortunately I cant open it, it keep sending me error!

Re: DCCGARCH11

Posted: Tue Jun 10, 2014 10:40 am
by trubador
PDF document should be stored here: C:\Users\...\Documents\EViews Addins\dccgarch11\dccgarch11.pdf

Re: DCCGARCH11

Posted: Wed Jun 11, 2014 6:45 am
by econworker
trubador wrote:PDF document should be stored here: C:\Users\...\Documents\EViews Addins\dccgarch11\dccgarch11.pdf


Thank you very much

Re: DCCGARCH11

Posted: Wed Jun 18, 2014 9:13 am
by akash27
Before estimating a DCC-GARCH(1,1) model, time series have to be filtered to assure zero expected (mean) value of the time series. Usually, a bivariate Vector Autoregressive (VAR) model used to initially remove potential linear structure, then the residuals of the VAR model are used as inputs for the DCCGARCH model.
Is this procedure involved in the add-in or do I need to do this separately and give residuals as input to the add-in?

Re: DCCGARCH11

Posted: Wed Jun 18, 2014 11:32 am
by akash27
After running the add-in, how do I get graphical plot of DCC GARCH conditional correlation ( say on y-axis) with time (say on x-axis) between two series?

Re: DCCGARCH11

Posted: Thu Jun 19, 2014 2:34 am
by trubador
If your questions are regarding the general use of the add-in, just post them under the relevant thread in this forum. Please do not send private e-mail messages (especially not to my office account).

akash27 wrote:Before estimating a DCC-GARCH(1,1) model, time series have to be filtered to assure zero expected (mean) value of the time series. Usually, a bivariate Vector Autoregressive (VAR) model used to initially remove potential linear structure, then the residuals of the VAR model are used as inputs for the DCCGARCH model.Is this procedure involved in the add-in or do I need to do this separately and give residuals as input to the add-in?

The add-in only allows univariate AR-X models for mean equations. VAR model needs to be specified at the outset.
akash27 wrote:After running the add-in, how do I get graphical plot of DCC GARCH conditional correlation ( say on y-axis) with time (say on x-axis) between two series?

If you select "Save dynamic correlations" option, then you can plot either rho_??_?? for the series of interest or rhos?? for all correlations post estimation. For instance, for a two-variable model it would be:

Code: Select all

line rho_12_01
line rhos01