Answers are grounded in this paper and cite their sources.
This paper deals with recovering an unknown vector β from the noisy data Y = Xβ + σξ, where X is a known n × p matrix with n ≥ p and ξ is a standard white Gaussian noise. In order to estimate β, a spectral cut-off estimate β ᵐ (Y) with a data-driven cut-off frequency m(Y) is used. The cut-off frequency is selected as a minimizer of the unbiased risk estimate of the mean square prediction error, i.e., m = min ₘ \ \| Y - X β ᵐ ( Y ) \|² + 2σ ² m \ . Assuming that β belongs to an ellipsoid W, we derive upper bounds for the maximal risk β ∈ W E\| β ᵐ ( Y ) - β \|² and show that β ᵐ ( Y ) is a rate optimal minimax estimator overW.
No takes yet. Share an insight, caveat, or question.
Chernousova et al. (2014) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: