New methods for variable importance testing with applications to genetic studies
Abstract/Contents
- Abstract
- The objective of this thesis is to develop new practical and principled statistical methodology for the analysis of genome-wide association data, in order to identify, as precisely as possible, the genetic variants that affect complex phenotypes. This problem can be stated as one of testing multiple hypotheses of conditional independence between many possible explanatory variables and a response of interest, within a high-dimensional non-parametric regression setting. This dissertation builds upon previous work on knockoffs, which provides a general framework for addressing such variable importance testing problems. In particular, we study how to generate valid knockoffs for genetic variants, while taking into account the particular structure of these data and the hidden Markov models developed by geneticists to describe their distribution. As a result, we can obtain an effective and statistically rigorous tool for genetic mapping that controls the false discovery rate under minimal assumptions, while overcoming many of the limitations of the existing state-of-the-art methods. Extensive numerical experiments with genetic data confirm the empirical validity and effectiveness of our method, while applications to the analysis of large data sets lead to many new discoveries
Description
Type of resource | text |
---|---|
Form | electronic resource; remote; computer; online resource |
Extent | 1 online resource |
Place | California |
Place | [Stanford, California] |
Publisher | [Stanford University] |
Copyright date | 2020; ©2020 |
Publication date | 2020; 2020 |
Issuance | monographic |
Language | English |
Creators/Contributors
Author | Sesia, Matteo |
---|---|
Degree supervisor | Candès, Emmanuel J. (Emmanuel Jean) |
Thesis advisor | Candès, Emmanuel J. (Emmanuel Jean) |
Thesis advisor | Sabatti, Chiara |
Thesis advisor | Tibshirani, Robert |
Degree committee member | Sabatti, Chiara |
Degree committee member | Tibshirani, Robert |
Associated with | Stanford University, Department of Statistics. |
Subjects
Genre | Theses |
---|---|
Genre | Text |
Bibliographic information
Statement of responsibility | Matteo Sesia |
---|---|
Note | Submitted to the Department of Statistics |
Thesis | Thesis Ph.D. Stanford University 2020 |
Location | electronic resource |
Access conditions
- Copyright
- © 2020 by Matteo Sesia
- License
- This work is licensed under a Creative Commons Attribution Non Commercial 3.0 Unported license (CC BY-NC).
Also listed in
Loading usage metrics...