Approaches for local false discovery rates
Abstract/Contents
- Abstract
- Modern hypothesis testing problems involve a large collection of hypotheses H_1, ..., H_N. The Bayesian local false discovery rate is the posterior probability that a hypothesis is null. The local false discovery rate paradigm is therefore an intuitive and attractive approach for deciding which hypotheses to declare as non-null. Accurate local false discovery rate estimation relies on positing an appropriate null distribution, which should be empirically estimated. It is often reasonable to estimate the null distribution from data that is truncated over a region which can safely deemed to be null. This truncation approach is appealing since it enriches the class of models that can be used as candidate null distributions. We propose the use of a log-concave null distribution as a non-parametric extension of earlier local false discovery rate work. Moreover, the truncation approach has a missing data interpretation. For exponential families, a general but simple EM algorithm can be constructed to obtain the maximum likelihood estimates without having to solve the corresponding score equations, which may be awkward in the case of truncation. We verify the correctness of the algorithm by applying Louis' formula (1982).
Description
| Type of resource | text |
|---|---|
| Form | electronic; electronic resource; remote |
| Extent | 1 online resource. |
| Publication date | 2013 |
| Issuance | monographic |
| Language | English |
Creators/Contributors
| Associated with | Angeles, Rudolfo Sombillo |
|---|---|
| Associated with | Stanford University, Department of Statistics. |
| Primary advisor | Holmes, Susan, 1954- |
| Thesis advisor | Holmes, Susan, 1954- |
| Thesis advisor | Efron, Bradley |
| Thesis advisor | Walther, Guenther |
| Advisor | Efron, Bradley |
| Advisor | Walther, Guenther |
Subjects
| Genre | Theses |
|---|
Bibliographic information
| Statement of responsibility | Rudolfo Sombillo Angeles. |
|---|---|
| Note | Submitted to the Department of Statistics. |
| Thesis | Thesis (Ph.D.)--Stanford University, 2013. |
| Location | electronic resource |
Access conditions
- Copyright
- © 2013 by Rudolfo Sombillo Angeles
- 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...