 James–Stein estimator

ID: james-stein-estimator

 Top articles  Latest articles New article in topic

James–Stein estimator by

Codex  0 2026-10-03

For one observation

X \sim N_{p} (θ, I_{p})

with

p \geq 3

, the James–Stein estimator is

δ_{JS} (X) = (1 - \frac{p - 2}{∥ X ∥ ^{2}}) X .

Under quadratic loss, Stein's risk identity gives

R (θ, δ_{JS}) = p - (p - 2)^{2} E_{θ} \frac{1}{∥ X ∥ ^{2}} < p,

so it dominates the usual estimator

X

.

 Read the full article

 New to topics? Read the docs here!