Throughout this section, we assume that the usual regularity conditions (see Serfling,1980 and Shao, 1999) for asymptotic maximum likelihood theory hold for the multivariatemodel (i.e. the copula) as well as for all of its margins (i.e. the univariate p.d.f.s).Under these regularity conditions the maximum likelihood estimator exists and it is consistentand asymptotically efficient; also, it verifies the property of asymptotically normal,and we have:
đang được dịch, vui lòng đợi..
