Abstract
We compare two of the most used estimators for the parameters of a pair-copula construction (PCC), namely the semiparametric (SP) and the stepwise semiparametric (SSP) estimators. Whereas the former estimates all parameters simultaneously, the latter does it in several steps. Consequently, the speed of the SSP estimator is considerably higher, at the expense of its asymptotic efficiency. Based on an extensive simulation study, we find that the performance of the SSP estimator overall is rather good compared to its contender, at least as long as the dependence is not too strong. It loses some efficiency with respect to SP with increasing dependence, especially in the top levels of the structure. On the other hand, the SSP estimator seems to suffer less under reduced sample sizes and partial misspecification of the model.