On the limitations of Barankin type bounds for MLE threshold prediction

Koby Todros; Roni Winik; Joseph Tabrikian

doi:10.1016/j.sigpro.2014.10.016

On the limitations of Barankin type bounds for MLE threshold prediction

Koby Todros, Roni Winik, Joseph Tabrikian

Department of Electrical & Computer Engineering

Ben-Gurion University of the Negev

Research output: Contribution to journal › Article › peer-review

Abstract

Various approximations of the Barankin bound (BB), called Barankin type bounds (BTBs), are usually used to predict the threshold region of the maximum likelihood estimator (MLE). In this paper, we show that BTBs are not always appropriate for predicting this threshold region. We prove that in some common estimation problems the BB is infinite, and thus, its approximations cannot reliably predict the MLE threshold region. This result is illustrated for two general signal processing models, in which BTBs can become arbitrarily large, and consequently, cannot reliably predict the MLE threshold region.

Original language	American English
Pages (from-to)	622-627
Number of pages	6
Journal	Signal Processing
Volume	108
DOIs	https://doi.org/10.1016/j.sigpro.2014.10.016
State	Published - 1 Jan 2015

Keywords

Barankin bound
Performance bounds
Periodic parameter estimation
Threshold phenomenon
Uniformly unbiased estimators

All Science Journal Classification (ASJC) codes

Control and Systems Engineering
Software
Signal Processing
Computer Vision and Pattern Recognition
Electrical and Electronic Engineering

Access to Document

10.1016/j.sigpro.2014.10.016

Cite this

@article{5437bcc2aa2f42dcae9feac51f2838dd,

title = "On the limitations of Barankin type bounds for MLE threshold prediction",

abstract = "Various approximations of the Barankin bound (BB), called Barankin type bounds (BTBs), are usually used to predict the threshold region of the maximum likelihood estimator (MLE). In this paper, we show that BTBs are not always appropriate for predicting this threshold region. We prove that in some common estimation problems the BB is infinite, and thus, its approximations cannot reliably predict the MLE threshold region. This result is illustrated for two general signal processing models, in which BTBs can become arbitrarily large, and consequently, cannot reliably predict the MLE threshold region.",

keywords = "Barankin bound, Performance bounds, Periodic parameter estimation, Threshold phenomenon, Uniformly unbiased estimators",

author = "Koby Todros and Roni Winik and Joseph Tabrikian",

note = "Publisher Copyright: {\textcopyright} 2014 Elsevier B.V.",

year = "2015",

month = jan,

day = "1",

doi = "10.1016/j.sigpro.2014.10.016",

language = "American English",

volume = "108",

pages = "622--627",

journal = "Signal Processing",

issn = "0165-1684",

publisher = "Elsevier B.V.",

}

TY - JOUR

T1 - On the limitations of Barankin type bounds for MLE threshold prediction

AU - Todros, Koby

AU - Winik, Roni

AU - Tabrikian, Joseph

PY - 2015/1/1

Y1 - 2015/1/1

N2 - Various approximations of the Barankin bound (BB), called Barankin type bounds (BTBs), are usually used to predict the threshold region of the maximum likelihood estimator (MLE). In this paper, we show that BTBs are not always appropriate for predicting this threshold region. We prove that in some common estimation problems the BB is infinite, and thus, its approximations cannot reliably predict the MLE threshold region. This result is illustrated for two general signal processing models, in which BTBs can become arbitrarily large, and consequently, cannot reliably predict the MLE threshold region.

AB - Various approximations of the Barankin bound (BB), called Barankin type bounds (BTBs), are usually used to predict the threshold region of the maximum likelihood estimator (MLE). In this paper, we show that BTBs are not always appropriate for predicting this threshold region. We prove that in some common estimation problems the BB is infinite, and thus, its approximations cannot reliably predict the MLE threshold region. This result is illustrated for two general signal processing models, in which BTBs can become arbitrarily large, and consequently, cannot reliably predict the MLE threshold region.

KW - Barankin bound

KW - Performance bounds

KW - Periodic parameter estimation

KW - Threshold phenomenon

KW - Uniformly unbiased estimators

UR - http://www.scopus.com/inward/record.url?scp=84919338486&partnerID=8YFLogxK

U2 - 10.1016/j.sigpro.2014.10.016

DO - 10.1016/j.sigpro.2014.10.016

M3 - Article

SN - 0165-1684

VL - 108

SP - 622

EP - 627

JO - Signal Processing

JF - Signal Processing

ER -

On the limitations of Barankin type bounds for MLE threshold prediction

Abstract

Keywords

All Science Journal Classification (ASJC) codes

Access to Document

Other files and links

Fingerprint

Cite this