Confused about the unconditional covariance matrix in a DCC GARCH model. Could anyone help me understand it? My understanding is that we get the unconditional covariance before based on the data sets. For example, two data sets, A and B, then the unconditional covariance matrix is built by the variance of A and B respectively and covariance of them, is that true? Thanks.

  • $\begingroup$ Do you have a link to what you are reading about DCC GARCH at the moment? $\endgroup$
    – Richi Wa
    Apr 1, 2015 at 14:27
  • $\begingroup$ Yes, @Richard. [link](pages.stern.nyu.edu/~rengle/Dcc-Sheppard.pdf ). Words to explain the unconditional covariance is on the page 5 under equation (2). Thanks for you reply. $\endgroup$
    – Fly_back
    Apr 1, 2015 at 16:27

1 Answer 1


Your question is about the $\overline{Q}$, right?

If so, it is the covariance between the error terms $E_{i,t}$ and $E_{j,t}$.(The sub term $i$ comes from asset $i$, and $j$ for asset $j$)

  • $\begingroup$ Please be straight to the point when answering, it's just clearer for everybody. $\endgroup$
    – SRKX
    Apr 10, 2015 at 5:01
  • $\begingroup$ Thanks @SRKX. I understand it now. So one more question, $\bar{Q}$ can be estimated by the standard residuals, so is this matrix is the parameters in the model, i n other words, I want to estimate the Bayesian Information Criterion (BIC), and I need to know the number of parameters, I wonder whether the three parameters in the matrix needs to be considered. If not considered, then there will be 8 parameter in a bivariate case, otherwise, there will be 11 parameters. $\endgroup$
    – Fly_back
    May 14, 2015 at 14:47
  • $\begingroup$ Dear @SRKX, sorry for the previous question, I have already understood the model now. The elements in $\bar{Q}$ should be considered in. $\endgroup$
    – Fly_back
    May 14, 2015 at 16:38

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

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