Breakdown points of trimmed likelihood estimators and related estimators in generalized linear models
Posted by dhin pada 26 April 2009
Breakdown points of trimmed likelihood estimators and related estimators
in generalized linear models
By CHRISTINE H. MULLER and NEYKO NEYKOV March 2001
Lower bounds for breakdown points of trimmed likelihood (TL) estimators in a general setup are expressed by the fullness parameter of Vandev (1993), where results of Vandev and Neykov (1998) are extended. A special application of the general result are the breakdown points of TL estimators and related estimators as the S estimators in generalized linear models. For the generalized linear models, a connection between the fullness parameter and the quantity N(X) of Muller (1995) is derived for the case that the explanatory variables may be not in general position which happens in particular in designed experiments. These results are in particular applied to logistic regression and log-linear models where also upper bounds for the breakdown points are derived.