Robust estimators for functional logistic regression models based on $B-$splines

BOENTE, GRACIELA; VALDORA, MARINA

Toulouse

Conferencia; International Conference on Robust Statistics (ICORS 2023); 2023

Toulouse School of Economics - TSE

Taking into account the sensitivity of these estimators to atypical observations andbased on the ideas given for euclidean covariates by Bianco & Yohai [1996] andCroux & Haesbroeck [2003], we define robust estimators of the intercept 0 and theslope 0 following a sieve approach combined with weighted M−estimators. Theoreticalassurances regarding the consistency and convergence rates of our proposalwill be presented. Besides, through the results of a numerical study, we will illustratethe sensitivity of the classical estimator and the stability of the proposed method.