A1155
Title: Robust and efficient copulas: From moment redundancy to Neyman orthogonality
Authors: Artem Prokhorov - University of Sydney (Australia) [presenting]
Abstract: Consider a general multivariate model where univariate marginal distributions are known up to a parameter vector and we are interested in estimating that parameter vector without specifying the joint distribution, except for the marginals. If we assume independence between the marginals and maximize the resulting quasi-likelihood, we obtain a consistent but inefficient QMLE. If we assume a parametric copula (other than independence) we obtain a full MLE, which is efficient but only under a correct copula specification and is biased otherwise. The Holy Grail of dependence modelling is to characterize a copula function that improves efficiency without sacrificing robustness. The talk is about a few parametric and semiparametric results obtained in search of the holy grail in recent years.
