Robustness in Semiparametric Partly Linear Models

BIANCO, ANA; BOENTE, GRACIELA

Ambreres, Belgica

Congreso; International Conference on Robust Statistics (ICORS 2003); 2003

In this talk, we will first remind the classical approach to partly linear autoregressive models. The sensitivity to outliers of the classical estimates for these models is good evidence that robust methods are needed. The problem of obtaining a family of robust estimates, in a partly linear autoregression model, is then addressed introducing a three–step robust procedure whose asymptotic behavior is derived. A robust procedure to choose the smoothing parameter is also discussed. Through a Monte Carlo study, the performance of the proposed estimates is compared with the classical ones, under the partly linear autoregressive model (1). Moreover, a procedure to detect anomalous observations is also discussed.