A general framework for estimation and inference from prototypes of feature clusters
Abstract/Contents
- Abstract
- Presented with a dataset with predefined variable grouping structure, I propose using response-aware prototypes (supervised by the response) in further analysis of the data. A framework for testing group-wide signal amongst variables is presented and developed. It is shown that tests based on supervised prototypes are more powerful than competitors from the classical literature. Furthermore, various aspects of multivariate modeling (estimation, prediction and variable selection) on prototypes are considered. A new penalty function for linear regression is proposed, meant to be used when predictions from various datasets need to be combined. Also, a procedure for generating highly interpretable snapshots of the dataset is presented and demonstrated. Finally, FDR controlling knockoff procedures are extended to the prototype level. Overall, prototyping is shown to be a versatile tool for data analysis. Since prototypes are constructed with reference to the response, care needs to be taken to account for this properly in subsequent analysis. Modern statistical inferential tools from the recently developed selective inference literature are applied to this problem throughout.
Description
Type of resource | text |
---|---|
Form | electronic; electronic resource; remote |
Extent | 1 online resource. |
Publication date | 2016 |
Issuance | monographic |
Language | English |
Creators/Contributors
Associated with | Reid, Stephen |
---|---|
Associated with | Stanford University, Department of Statistics. |
Primary advisor | Tibshirani, Robert |
Thesis advisor | Tibshirani, Robert |
Thesis advisor | Hastie, Trevor |
Thesis advisor | Taylor, Jonathan E |
Advisor | Hastie, Trevor |
Advisor | Taylor, Jonathan E |
Subjects
Genre | Theses |
---|
Bibliographic information
Statement of responsibility | Stephen Reid. |
---|---|
Note | Submitted to the Department of Statistics. |
Thesis | Thesis (Ph.D.)--Stanford University, 2016. |
Location | electronic resource |
Access conditions
- Copyright
- © 2016 by Stephen Reid
- 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...