Javascript must be enabled for the correct page display

Sparse Inference of High Dimensional Survival Models

Velt, V (2016) Sparse Inference of High Dimensional Survival Models. Master's Thesis / Essay, Mathematics.

[img]
Preview
Text
Sparse_Inference_of_High_Dimen_1.pdf - Published Version

Download (1MB) | Preview
[img] Text
toestemming.pdf - Other
Restricted to Backend only

Download (496kB)

Abstract

This thesis concerns sparse inference of high dimensional survival models. By describing the statistical model space as a Riemannian manifold, we discuss the geometrical structure of generalized linear models. Then we use this geometrical interpretation to find a sparse solution path that leads to the maximum likelihood estimator. The method we suggest to find such a sparse solution path, is based on differential geometric least angle regression. For the purpose of this thesis the method is extended to excess relative risk survival models. The derived method is implemented in R and used for a data analysis on data of a survival cancer study. In the analysis, a genomic signature is found that might be predictive for the prospects of an individual suffering from a specific type of ovarian

Item Type: Thesis (Master's Thesis / Essay)
Degree programme: Mathematics
Thesis type: Master's Thesis / Essay
Language: English
Date Deposited: 15 Feb 2018 08:10
Last Modified: 15 Feb 2018 08:10
URI: https://fse.studenttheses.ub.rug.nl/id/eprint/13646

Actions (login required)

View Item View Item