我也很想知道，楼主现在有答案么？给我说说？

用S-PLUS软件可以实现

thanks a lot

同求，等大神赐教！

open data C:\Users\Administrator\Desktop\china-usd-ret.xls
data(format=xls,org=columns) 2 869 china usd
set china = china
set usd = usd
*VAR(1) model for the mean, BEKK for the variance
system(model=var1)
variables china usd
lags 1
det constant
end(system)
garch(p=2,q=3,model=var1,distrib=t,mv=bekk,pmethod=bfgs,piters=20,robust,hmatrices=hh,rvectors=rr) / china usd


set resid1 %regstart() %regend() = rr(t)(1)
set resid2 %regstart() %regend() = rr(t)(2)
set variance1 %regstart() %regend() = hh(t)(1,1)
set variance2 %regstart() %regend() =hh(t)(2,2)
set cov12 %regstart() %regend() = hh(t)(1,2)

set stdxjpn = rr(t)(1)/sqrt(hh(t)(1,1))
set stdxfra = rr(t)(2)/sqrt(hh(t)(2,2))

@mvqstat(lags=12)
# stdxjpn
@mvqstat(lags=12)
# stdxfra
@mvqstat(lags=12)
# stdxjpn  stdxfra

*****
set stdxjpnsq = stdxjpn**2
set stdxfrasq = stdxfra**2
@mvqstat(lags=12)
# stdxjpnsq
@mvqstat(lags=12)
# stdxfrasq
@mvqstat(lags=12)
# stdxjpnsq  stdxfrasq
例子用的winrats软件