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.
Jan JOHANNES Aug 04,2023 Theses
Thesis:
Bachelor in Mathematics

Author:
Dominik Daniel

Title:
Nonparametric survival function estimation for censored data

Supervisor:
Jan JOHANNES

Abstract:
In this bachelor thesis we consider a strategy for estimating the survival function as presented in [BBC18], associated to an event time determined using censored interval data. The regression analysis is based on the method of least squares where the parameters correspond to the coefficients arising from the expansion of S for the chosen orthonormal basis. For bases with compact support, we get adaptive results which lead to general non- parametric rates. The results can be used for bases with no compact support which is new for regression models. Here, the Laguerre basis is used whose support lies on R+ and is therefore well suited if there are non-negative random variables in the model. Simulation results have shown that their estimator works well, especially in very general contexts. At the end of this work, we consider a simulation with a self-implemented estimator.

References:
O. Bouaziz, E. Brunel, and F. Comte. Nonparametric survival function estimation for data subject to interval censoring case 2, MAP5 2018-11. 2018 hal-01766456.