T. Hobza, L. Pardo Llorente, N. Martín
This contribution deals with robust estimation of parameters of logistic regression models. It introduces a modified median estimator of the underlying parameters of these models based on statistically smoothed binary responses. Sensitivity to contaminations and leverage points is studied by simulations and compared in this manner with the sensitivity of some robust estimators previously introduced to the logistic regression. The new estimator appears to be more robust for larger sample sizes and higher levels of contamination.
Palabras clave: Logistic regression, Median estimator, Robustness
Programado
X05.3 Inferencia Estadística I
7 de septiembre de 2016 12:30
Aula 21.08
