Page 1 of 1
Q stat DCC GARCH
Posted: Wed Feb 08, 2012 3:18 pm
by aymenbelgacem
Hi
I'm using an old version of Engle(2002) JBES for the DCC GARCH and I would like to calculate the MV Qstat
Is there any suggestion plz.
I don't use the GARCH instruction to estimate the model because I would like to introduce some other variables in GARCH and Q equations...
This is a part of the code
Thanks
Aymen
Re: Q stat DCC GARCH
Posted: Wed Feb 08, 2012 5:56 pm
by TomDoan
Do you mean something other than the @MVQSTAT procedure?
Re: Q stat DCC GARCH
Posted: Wed Feb 08, 2012 8:50 pm
by aymenbelgacem
Yes I mean the @mvqstat but I don't know how to treat the dynamic correlation before using the procedure
Thanks Tom
Re: Q stat DCC GARCH
Posted: Thu Feb 09, 2012 9:10 am
by TomDoan
The GARCHMV.RPF example includes multivariate standardization of the residuals prior to calling @MVQSTAT:
Code: Select all
garch(p=1,q=1,pmethod=simplex,piters=10,$
hmatrices=hh,rvectors=rd) / xjpn xfra xsui
set z1 = rd(t)(1)/sqrt(hh(t)(1,1))
set z2 = rd(t)(2)/sqrt(hh(t)(2,2))
set z3 = rd(t)(3)/sqrt(hh(t)(3,3))
@bdindtests(number=40) z1
@bdindtests(number=40) z2
@bdindtests(number=40) z3
*
* Multivariate Q statistic. This requires transforming the residuals to
* eliminate the time-varying correlations.
*
dec vect[series] zu(%nvar)
do time=%regstart(),%regend()
compute %pt(zu,time,%solve(%decomp(hh(time)),rd(time)))
end do time
@mvqstat(lags=40)
# zu
Re: Q stat DCC GARCH
Posted: Thu Feb 09, 2012 9:42 am
by aymenbelgacem
Thanks Tom
I know this example but I can't use the GARCH instruction and the Xregressors because I want to add the lagged variance of the market i in the market j and other variables, so I tried to paste the second part of GRACHMV example in which one calculates the standardized residuals. But the problem that I tried to make a link between it and my version of the DCC GARCH attached above, I didn't find what corresponds to "hh" and "rd" in my old version of the DCC.
Sorry Tom for this (naive) question
Re: Q stat DCC GARCH
Posted: Fri Feb 10, 2012 7:51 am
by TomDoan
Your equivalent would be to use %XT for the HH and U arrays of series:
Code: Select all
dec vect[series] zu(n)
do time=%regstart(),%regend()
compute %pt(zu,time,%solve(%decomp(%xt(hh,time)) ,%xt(u,time)))
end do time
Re: Q stat DCC GARCH
Posted: Fri Feb 10, 2012 8:55 am
by aymenbelgacem
I don't have HH. Do you mean h?
Re: Q stat DCC GARCH
Posted: Fri Feb 10, 2012 11:07 am
by moderator
It would be whatever name you are using for your variances and covariances in your particular code. Mr. Doan used HH in his example, but it looks like you used H in your code.
Regards,
Tom Maycock