Estimation of Bounded Normal Mean: An Alternative Proof for the Discreteness of the Least Favorable Prior

Semih Yagli; Alex Dytso; H. Vincent Poor

doi:10.1109/ITW44776.2019.8988927

Estimation of Bounded Normal Mean: An Alternative Proof for the Discreteness of the Least Favorable Prior

Semih Yagli, Alex Dytso, H. Vincent Poor

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

2 Scopus citations

Abstract

This paper studies the classical Bayesian normal mean estimation problem where the estimand is assumed to be contained in a bounded set. It is known that the least favorable distribution for this mean estimation problem is discrete with finitely many mass points. This work offers an alternative proof utilizing the variational diminishing property of Gaussian kernels.

Original language	English (US)
Title of host publication	2019 IEEE Information Theory Workshop, ITW 2019
Publisher	Institute of Electrical and Electronics Engineers Inc.
ISBN (Electronic)	9781538669006
DOIs	https://doi.org/10.1109/ITW44776.2019.8988927
State	Published - Aug 2019
Event	2019 IEEE Information Theory Workshop, ITW 2019 - Visby, Sweden Duration: Aug 25 2019 → Aug 28 2019

Publication series

Name	2019 IEEE Information Theory Workshop, ITW 2019

Conference

Conference	2019 IEEE Information Theory Workshop, ITW 2019
Country/Territory	Sweden
City	Visby
Period	8/25/19 → 8/28/19

All Science Journal Classification (ASJC) codes

Software
Computational Theory and Mathematics
Computer Networks and Communications
Information Systems

Access to Document

10.1109/ITW44776.2019.8988927

Cite this

Yagli, S., Dytso, A., & Poor, H. V. (2019). Estimation of Bounded Normal Mean: An Alternative Proof for the Discreteness of the Least Favorable Prior. In 2019 IEEE Information Theory Workshop, ITW 2019 Article 8988927 (2019 IEEE Information Theory Workshop, ITW 2019). Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/ITW44776.2019.8988927

@inproceedings{bd736954e7bc4400af6b0878365ffb16,

title = "Estimation of Bounded Normal Mean: An Alternative Proof for the Discreteness of the Least Favorable Prior",

abstract = "This paper studies the classical Bayesian normal mean estimation problem where the estimand is assumed to be contained in a bounded set. It is known that the least favorable distribution for this mean estimation problem is discrete with finitely many mass points. This work offers an alternative proof utilizing the variational diminishing property of Gaussian kernels.",

author = "Semih Yagli and Alex Dytso and Poor, {H. Vincent}",

year = "2019",

month = aug,

doi = "10.1109/ITW44776.2019.8988927",

language = "English (US)",

series = "2019 IEEE Information Theory Workshop, ITW 2019",

publisher = "Institute of Electrical and Electronics Engineers Inc.",

booktitle = "2019 IEEE Information Theory Workshop, ITW 2019",

address = "United States",

note = "2019 IEEE Information Theory Workshop, ITW 2019 ; Conference date: 25-08-2019 Through 28-08-2019",

}

Yagli, S, Dytso, A & Poor, HV 2019, Estimation of Bounded Normal Mean: An Alternative Proof for the Discreteness of the Least Favorable Prior. in 2019 IEEE Information Theory Workshop, ITW 2019., 8988927, 2019 IEEE Information Theory Workshop, ITW 2019, Institute of Electrical and Electronics Engineers Inc., 2019 IEEE Information Theory Workshop, ITW 2019, Visby, Sweden, 8/25/19. https://doi.org/10.1109/ITW44776.2019.8988927

Estimation of Bounded Normal Mean: An Alternative Proof for the Discreteness of the Least Favorable Prior. / Yagli, Semih; Dytso, Alex; Poor, H. Vincent.
2019 IEEE Information Theory Workshop, ITW 2019. Institute of Electrical and Electronics Engineers Inc., 2019. 8988927 (2019 IEEE Information Theory Workshop, ITW 2019).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

TY - GEN

T1 - Estimation of Bounded Normal Mean

T2 - 2019 IEEE Information Theory Workshop, ITW 2019

AU - Yagli, Semih

AU - Dytso, Alex

AU - Poor, H. Vincent

PY - 2019/8

Y1 - 2019/8

N2 - This paper studies the classical Bayesian normal mean estimation problem where the estimand is assumed to be contained in a bounded set. It is known that the least favorable distribution for this mean estimation problem is discrete with finitely many mass points. This work offers an alternative proof utilizing the variational diminishing property of Gaussian kernels.

AB - This paper studies the classical Bayesian normal mean estimation problem where the estimand is assumed to be contained in a bounded set. It is known that the least favorable distribution for this mean estimation problem is discrete with finitely many mass points. This work offers an alternative proof utilizing the variational diminishing property of Gaussian kernels.

UR - http://www.scopus.com/inward/record.url?scp=85081110415&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85081110415&partnerID=8YFLogxK

U2 - 10.1109/ITW44776.2019.8988927

DO - 10.1109/ITW44776.2019.8988927

M3 - Conference contribution

T3 - 2019 IEEE Information Theory Workshop, ITW 2019

BT - 2019 IEEE Information Theory Workshop, ITW 2019

PB - Institute of Electrical and Electronics Engineers Inc.

Y2 - 25 August 2019 through 28 August 2019

ER -

Estimation of Bounded Normal Mean: An Alternative Proof for the Discreteness of the Least Favorable Prior

Abstract

Publication series

Conference

All Science Journal Classification (ASJC) codes

Access to Document

Other files and links

Fingerprint

Cite this