de.mpg.escidoc.pubman.appbase.FacesBean
English
 
Help Guide Disclaimer Contact us Login
  Advanced SearchBrowse

Item

ITEM ACTIONSEXPORT

Released

Report

Bayesian Estimators for Robins-Ritov’s Problem

MPS-Authors
http://pubman.mpdl.mpg.de/cone/persons/resource/persons83954

Harmeling,  S
Department Empirical Inference, Max Planck Institute for Biological Cybernetics, Max Planck Society;

Locator
There are no locators available
Fulltext (public)
There are no public fulltexts available
Supplementary Material (public)
There is no public supplementary material available
Citation

Harmeling, S.(2007). Bayesian Estimators for Robins-Ritov’s Problem (EDI-INF-RR-1189).


Cite as: http://hdl.handle.net/11858/00-001M-0000-0013-CB97-8
Abstract
Bayesian or likelihood-based approaches to data analysis became very popular in the 64257;eld of Machine Learning. However, there exist theoretical results which question the general applicability of such approaches; among those a result by Robins and Ritov which introduce a speci64257;c example for which they prove that a likelihood-based estimator will fail (i.e. it does for certain cases not converge to a true parameter estimate, even given in64257;nite data). In this paper we consider various approaches to formulate likelihood-based estimators in this example, basically by considering various extensions of the presumed generative model of the data. We can derive estimators which are very similar to the classical Horvitz-Thompson and which also account for a priori knowledge of an observation probability function.