A new lower bound on the mean-square error of biased estimators

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

Abstract

In this paper, the class of lower bounds on the MSE of unbiased estimators, derived in our previous work, is extended to the case of biased estimation. The proposed class is derived by projecting the estimation error on a Hilbert subspace of 2, which contains linear transformations of elements in the domain of an integral transform of the likelihood-ratio function. It is shown that some well known bounds can be derived from the proposed class by modifying the kernel of the integral transform. By decomposing the projection of the estimation error into bias-independent and bias-dependent components, the proposed class is minimized with respect to the bias function subject to a bounded 2-norm of the bias-dependent component. A new computationally manageable bound is derived from the proposed class using the kernel of the weighted Fourier transform. The bound is applied for exploring the bias-variance tradeoff in the problem of direction-of-arrival estimation.

Original languageEnglish
Title of host publication2008 IEEE 25th Convention of Electrical and Electronics Engineers in Israel, IEEEI 2008
Pages745-749
Number of pages5
DOIs
StatePublished - 1 Dec 2008
Event2008 IEEE 25th Convention of Electrical and Electronics Engineers in Israel, IEEEI 2008 - Eilat, Israel
Duration: 3 Dec 20085 Dec 2008

Publication series

NameIEEE Convention of Electrical and Electronics Engineers in Israel, Proceedings

Conference

Conference2008 IEEE 25th Convention of Electrical and Electronics Engineers in Israel, IEEEI 2008
Country/TerritoryIsrael
CityEilat
Period3/12/085/12/08

Keywords

  • Bias-variance tradeoff
  • Biased estimation
  • Mean-square-error bounds
  • Parameter estimation

ASJC Scopus subject areas

  • Electrical and Electronic Engineering

Fingerprint

Dive into the research topics of 'A new lower bound on the mean-square error of biased estimators'. Together they form a unique fingerprint.

Cite this