A0180
Title: Prediction and robustness
Authors: Setareh Ranjbar - HEC Lausanne (Switzerland)
Elvezio Ronchetti - University of Geneva (Switzerland) [presenting]
Stefan Sperlich - University of Geneva (Switzerland)
Abstract: A general discussion of the robustness issues in a prediction framework is provided, and their implications in different areas, including classification, insurance, and estimation in finite populations are analyzed. We illustrate more specifically these issues in the prediction of nonlinear indices (such as inequality or poverty measures) for small areas and in the presence of outliers. We propose two approaches to calibrate for the bias of nonlinear functionals, such as the Gini index and when the so called representative outliers come from a skewed heavy tail distribution.
