False discovery rate control for spatial data
Abstract/Contents
- Abstract
- In many modern applications the aim of the statistical analysis is to identify `interesting' or `differentially behaved' regions from noisy spatial measurements. From a statistical standpoint the task is both to identify a collection of regions which are likely to be non-null, and to associate to this collection a measure of uncertainty. Viewing this task as a large scale multiple testing problem, we present methods for controlling the clusterwise false discovery rate, defined as the expected fraction of reported regions that are in truth null. Our methods extend the recent work of Siegmund, Zhang, and Yakir (2011), and can be applied whenever the high level excursions of the noise process are well approximated by a (potentially inhomogeneous) Poisson process. Borrowing ideas from the Poisson clumping heuristic literature, we show that the widely used pointwise procedure generally fails to control the clusterwise FDR. We also present a general framework for incorporating various measures of cluster significance into the clusterwise false discovery control procedure. We show that incorporating cluster size can result in a significant increase in power.
Description
Type of resource | text |
---|---|
Form | electronic; electronic resource; remote |
Extent | 1 online resource. |
Publication date | 2014 |
Issuance | monographic |
Language | English |
Creators/Contributors
Associated with | Chouldechova, Alexandra |
---|---|
Associated with | Stanford University, Department of Statistics. |
Primary advisor | Candès, Emmanuel J. (Emmanuel Jean) |
Primary advisor | Tibshirani, Robert |
Thesis advisor | Candès, Emmanuel J. (Emmanuel Jean) |
Thesis advisor | Tibshirani, Robert |
Thesis advisor | Siegmund, David |
Advisor | Siegmund, David |
Subjects
Genre | Theses |
---|
Bibliographic information
Statement of responsibility | Alexandra Chouldechova. |
---|---|
Note | Submitted to the Department of Statistics. |
Thesis | Thesis (Ph.D.)--Stanford University, 2014. |
Location | electronic resource |
Access conditions
- Copyright
- © 2014 by Olexandra Chouldechova
- 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...