Saved in:
Bibliographic Details
Main Authors: Perry, Ronan, Xu, Zichun, McGough, Olivia, Witten, Daniela
Format: Preprint
Published: 2024
Subjects:
Online Access:https://arxiv.org/abs/2408.06323
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866917292911099904
author Perry, Ronan
Xu, Zichun
McGough, Olivia
Witten, Daniela
author_facet Perry, Ronan
Xu, Zichun
McGough, Olivia
Witten, Daniela
contents In recent years, there has been substantial interest in the task of selective inference: inference on a parameter that is selected from the data. Many of the existing proposals fall into what we refer to as the \emph{infer-and-widen} framework: they produce symmetric confidence intervals whose midpoints do not account for selection and therefore are biased; thus, the intervals must be wide enough to account for this bias. In this paper, we investigate infer-and-widen approaches in three vignettes: the winner's curse, maximal contrasts, and inference after the lasso. In each of these examples, we show that a state-of-the-art infer-and-widen proposal leads to confidence intervals that are wider than a non-infer-and-widen alternative. Furthermore, even an ``oracle'' infer-and-widen confidence interval -- the narrowest possible interval that could be theoretically attained via infer-and-widen -- can be wider than the alternative.
format Preprint
id arxiv_https___arxiv_org_abs_2408_06323
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Infer-and-widen, or not?
Perry, Ronan
Xu, Zichun
McGough, Olivia
Witten, Daniela
Methodology
In recent years, there has been substantial interest in the task of selective inference: inference on a parameter that is selected from the data. Many of the existing proposals fall into what we refer to as the \emph{infer-and-widen} framework: they produce symmetric confidence intervals whose midpoints do not account for selection and therefore are biased; thus, the intervals must be wide enough to account for this bias. In this paper, we investigate infer-and-widen approaches in three vignettes: the winner's curse, maximal contrasts, and inference after the lasso. In each of these examples, we show that a state-of-the-art infer-and-widen proposal leads to confidence intervals that are wider than a non-infer-and-widen alternative. Furthermore, even an ``oracle'' infer-and-widen confidence interval -- the narrowest possible interval that could be theoretically attained via infer-and-widen -- can be wider than the alternative.
title Infer-and-widen, or not?
topic Methodology
url https://arxiv.org/abs/2408.06323