Robust estimation in single index models with asymmetric errors

AGOSTINELLI, CLAUDIO; BIANCO, ANA; BOENTE, GRACIELA

Ginebra

Conferencia; International Conference on Robust Statistics (ICORS 2016); 2016

Universidad de Ginebra

A family of robust estimators based on a three{step procedure related to a profile approach is proposed. When nuisance parameters are present, they may be estimated using a preliminary S-estimator which will allow to define also the tuning constant. In particular, we have introduced a robust consistent estimator of the nuisance parameter for the usual regression model with symmetric errors and for the log-gamma regression model. Consistency results for the robust profile proposal is obtained. A preliminary simulation study is presented so as to validate the good behaviour of the estimators under the central model andunder different contaminated scenarios.