CMStatistics 2017: Start Registration
View Submission - CMStatistics
Title: Uniformly valid confidence intervals post-model-selection Authors:  Francois Bachoc - Universite Paul Sabatier (France)
David Preinerstorfer - Université libre de Bruxelles (Belgium) [presenting]
Lukas Steinberger - University of Vienna (Austria)
Abstract: General methods are suggested to construct asymptotically uniformly valid confidence intervals post-model-selection. The constructions are based on principles recently proposed in the literature. In particular, the candidate models used can be misspecified, the target of inference is model-specific, and coverage is guaranteed for any data-driven model selection procedure. After developing a general theory, we apply our methods to practically important situations where the candidate set of models, from which a working model is selected, consists of fixed design homoskedastic or heteroskedastic linear models, or of binary regression models with general link functions.