UNIGE document Doctoral Thesis
previous document  unige:131077  next document
add to browser collection

Robust predictive distribution and bias-calibration in Linear Models

Defense Thèse de doctorat : Univ. Genève, 2019 - GSEM 74 - 2019/12/20
Abstract This thesis focuses on the concept of predictive distributions and bias calibration. At first, an extension of the concept of predictive distributions under contamination is studied in the case of Generalized Linear Models. A sensitivity analysis of the impact of contamination on the predictive distribution is studied making use of the class of $M$-estimators. In a second step, based on the available literature on bias-calibrated estimation in linear regression, the bases to implement bias-calibration for predictive distributions are studied and developed. This development is based on the finite-sample setting and an important aspect of the reasoning behind this contribution is the distinction between representative and non-representative outliers. As a result of this distinction, the use of the bias-calibration approach allows to integrate information from representative outliers within the predictive distribution.
Keywords Predictive distributionBias-calibrationRobust statisticsLaplace approximation
URN: urn:nbn:ch:unige-1310776
Full text
Thesis (1.1 MB) - public document Free access
(ISO format)
BRANCA, Mattia. Robust predictive distribution and bias-calibration in Linear Models. Université de Genève. Thèse, 2019. doi: 10.13097/archive-ouverte/unige:131077 https://archive-ouverte.unige.ch/unige:131077

273 hits



Deposited on : 2020-02-20

Export document
Format :
Citation style :