Title :
Optimal signal estimation using cross-validation
Author :
Nowak, Robert D.
Author_Institution :
Dept. of Electr. Eng., Michigan State Univ., East Lansing, MI, USA
Abstract :
This letter develops an optimal, nonlinear estimator of a deterministic signal in noise. The methods of penalized least-squares and cross-validation (CV) balance the bias-variance tradeoff and lead to a closed form expression for the estimator. The estimator is simultaneously optimal in a "small-sample", predictive sum of squares sense and asymptotically optimal in the mean square sense.
Keywords :
least squares approximations; noise; parameter estimation; signal processing; asymptotically optimal; bias-variance tradeoff; closed form expression; cross-validation; deterministic signal; mean square sense; noise; optimal nonlinear estimator; penalized least-squares method; signal estimation; small-sample predictive sum of squares sense; Costs; Equations; Estimation; Least squares approximation; Least squares methods; Linear regression; Signal processing; Statistics; Subspace constraints; Vectors;
Journal_Title :
Signal Processing Letters, IEEE
