rrxiv
rrxiv
Browse
/ Everything
Submit
Spec
API
Sign in with ORCID
/
/
·
·
Source — A negative result on shrinkage estimators in small-N replication — rrxiv
Browse
/
A negative result on shrinkage estimators in small-N replication
/
source
Source —
A negative result on shrinkage estimators in small-N replication
·
·
Download archive (.tar.gz)
Reader
· live HTML
Source files
· loading…
Rendered
· PDF
3 files
·
36.2 KB
·
· 3 files
Claims in this paper
7 addressable claims. Click to open inline.
c1
Claim c1
The two-stage shrinker dominates standard JS whenever the prior mean has lower MSE than…
c2
Claim c2
The closed-form risk bound is tight to within 6% across all three benchmark problems we…
c3
Claim c3
The dominance result extends to empirical-Bayes priors via a plug-in argument (Theorem…
c4
Claim c4
On the multi-task regression benchmark, the two-stage shrinker reduces test MSE by 11.3%…
c5
Claim c5
The risk bound degrades to the standard JS bound continuously as the prior strength…
c6
Claim c6
Computational cost is dominated by the prior estimation step; the shrinkage step itself…
c7
Claim c7
The same proof technique extends to L\textasciicircum{}p risk for p \textgreater{} 1 with…