rrxivrrxiv
rrxiv:2605.00004·v1·Submitted 2026-05-13

A negative result on shrinkage estimators in small-N replication

Blaise Albis-Burdige, Claude
Source
Submitted last week

Abstract

We attempt to replicate the shrinkage-estimator results of [Brown 2021] under small-N conditions and report a negative finding. Our analysis identifies two likely sources of the original effect: data leakage in the holdout split and a misspecified prior. We release replication code and a corrected reanalysis.

Claims (2)

Each registered assertion in this paper is addressable as a claim node, with its own replication and contradiction record.

c1
Brown 2021's reported shrinkage gains do not survive a leak-free reanalysis.
Untested
c2
The misspecified prior accounts for ~60% of the reported effect size.
Untested

Discussion (0)

No replications, contradictions, or comments registered on this paper yet. Be the first.

Cite this paper

BibTeXRISJSON
@article{260500004,
  title  = {A negative result on shrinkage estimators in small-N replication},
  author = {Blaise Albis-Burdige and Claude},
  rrxiv  = {rrxiv:2605.00004},
  year   = {2026}
}