ProbXiv
Explorersign in
Problem archiveProblem record

Statement

Let X∼MVN(θ,Id)X \sim MVN(\theta, I_d), where IdI_d denotes the dd-dimensional identity matrix. Consider the problem of estimating θ\theta using estimators of the form X+h(X)X + h(X), where hh is in some class H\mathcal{H} of functions. By SURE, [\hat{r}_h = d + 2\nabla \cdot h(X) + |h(X)|^2] is an unbiased estimator of the risk associated with hh. Let \begin{align*} h^* &= \text{arg min} (h \in \mathcal{H}: \hat{r}_h) \ \hat{\theta}^* &= X + h^(X). \end{align} Find an unbiased estimator r∗∗r^{**} of the risk E(∥θ^∗−θ∥2)\mathbb{E}(\|\hat{\theta}^* - \theta\|^2) of θ^∗\hat{\theta}^*.

Record

Added
Links

Comments

Nothing has been published against this problem yet, and nothing has been said about it.

Sign in with an institutional address to take part in the discussion. Reading every thread stays open to everyone.

Sign in

Solve with an agent

Open the statement in a chat, with the problem and the ground rules already written into the prompt.

This opens a third-party site. Nothing is posted back to ProbXiv and nothing you write there is recorded here — what a model gives you is an attempt, which a person still has to check.