Univ. Heidelberg
Statistics Group   Institute for Mathematics   Faculty of Mathematics and Computer Science   University Heidelberg
Ruprecht-Karls-Universität Heidelberg Institute for Mathematics Statistics of inverse problems Research Group
german english french



Publications
Cooperations
Research projects
Events
Teaching
Completed theses
People
Contact


Last edited on
Oct 17, 2024 by JJ
.
Thesis:
Bachelor in Mathematics

Author:
Laura Lecinena Pastor

Title:
Rho-Schätzer: eine neue Schätzmehode

Supervisor:
Jan JOHANNES

Abstract:
In this thesis we consider the problem of creating an universal, optimal and robust estimator. Therefore we define the so called ρ-estimator, for which quadratic risk we calculate an upper bound first for countable and than for uncountable models. This gives us the first step to being able to proof their optimality and robustness. Furthermore, we bound the minimax risk for generall models and more precise for histogramtype models and VC-subgraphs. Finally, we study the connection between the ρ-estimator and the maximum-likelihood estimator, showcasing it exemplarily for histogramtype models.

References:
Y. Baraud, L. Birgé, and M. Sart. A new method for estimation and model selection: Rho-estimation, Inventiones mathematicae, 207(2):425–517, 2016.