Induced cores and their use in robust parametric estimation

The recently proposed induced cores of continuous probability distributions are briefly introduced and used for a generalization of the classical moment estimation method. The suggested core moment estimators and, if necessary, the Huber moment estimators, are "tailored" to the assumed distributions similarly as the maximum likelihood estimators, with which in some cases coincide. However. the generalized moment estimates are robust. They are in multidimensional cases unexpectedly simple and their asymptotic relative efficiences seem to be reasonably near to one.