Multiple testing of local maxima for detection of peaks on the (celestial) sphere

Dan Cheng, Valentina Cammarota, Yabebal Fantaye, Domenico Marinucci, Armin Schwartzman

Research output: Contribution to journalArticle

Abstract

We present a topological multiple testing scheme for detecting peaks on the sphere under isotropic Gaussian noise, where tests are performed at local maxima of the observed field filtered by the spherical needlet transform. Our setting is different from the standard Euclidean large domain asymptotic framework, yet highly relevant to realistic experimental circumstances for some important areas of application in astronomy, namely point-source detection in cosmic Microwave Background radiation (CMB) data. Motivated by this application, we shall focus on cases where a single realization of a smooth isotropic Gaussian random field on the sphere is observed, and a number of well-localized signals are superimposed on such background field. The proposed algorithms, combined with the Benjamini-Hochberg procedure for thresholding p-values, provide asymptotic control of the False Discovery Rate (FDR) and power consistency as the signal strength and the frequency of the needlet transform get large.

Original languageEnglish (US)
Pages (from-to)31-60
Number of pages30
JournalBernoulli
Volume26
Issue number1
DOIs
StatePublished - Jan 1 2020

Fingerprint

Multiple Testing
Transform
Gaussian Random Field
Astronomy
Gaussian Noise
Point Source
p-Value
Thresholding
Microwave
Euclidean
Radiation
Background
False
Standards
Framework

Keywords

  • CMB
  • False discovery rate
  • Gaussian random fields
  • Height distribution
  • Needlet transform
  • Overshoot distribution
  • Power
  • Sphere

ASJC Scopus subject areas

  • Statistics and Probability

Cite this

Cheng, D., Cammarota, V., Fantaye, Y., Marinucci, D., & Schwartzman, A. (2020). Multiple testing of local maxima for detection of peaks on the (celestial) sphere. Bernoulli, 26(1), 31-60. https://doi.org/10.3150/18-BEJ1068

Multiple testing of local maxima for detection of peaks on the (celestial) sphere. / Cheng, Dan; Cammarota, Valentina; Fantaye, Yabebal; Marinucci, Domenico; Schwartzman, Armin.

In: Bernoulli, Vol. 26, No. 1, 01.01.2020, p. 31-60.

Research output: Contribution to journalArticle

Cheng, D, Cammarota, V, Fantaye, Y, Marinucci, D & Schwartzman, A 2020, 'Multiple testing of local maxima for detection of peaks on the (celestial) sphere', Bernoulli, vol. 26, no. 1, pp. 31-60. https://doi.org/10.3150/18-BEJ1068
Cheng, Dan ; Cammarota, Valentina ; Fantaye, Yabebal ; Marinucci, Domenico ; Schwartzman, Armin. / Multiple testing of local maxima for detection of peaks on the (celestial) sphere. In: Bernoulli. 2020 ; Vol. 26, No. 1. pp. 31-60.
@article{8721f842066e417e9dd775783fb1d94c,
title = "Multiple testing of local maxima for detection of peaks on the (celestial) sphere",
abstract = "We present a topological multiple testing scheme for detecting peaks on the sphere under isotropic Gaussian noise, where tests are performed at local maxima of the observed field filtered by the spherical needlet transform. Our setting is different from the standard Euclidean large domain asymptotic framework, yet highly relevant to realistic experimental circumstances for some important areas of application in astronomy, namely point-source detection in cosmic Microwave Background radiation (CMB) data. Motivated by this application, we shall focus on cases where a single realization of a smooth isotropic Gaussian random field on the sphere is observed, and a number of well-localized signals are superimposed on such background field. The proposed algorithms, combined with the Benjamini-Hochberg procedure for thresholding p-values, provide asymptotic control of the False Discovery Rate (FDR) and power consistency as the signal strength and the frequency of the needlet transform get large.",
keywords = "CMB, False discovery rate, Gaussian random fields, Height distribution, Needlet transform, Overshoot distribution, Power, Sphere",
author = "Dan Cheng and Valentina Cammarota and Yabebal Fantaye and Domenico Marinucci and Armin Schwartzman",
year = "2020",
month = "1",
day = "1",
doi = "10.3150/18-BEJ1068",
language = "English (US)",
volume = "26",
pages = "31--60",
journal = "Bernoulli",
issn = "1350-7265",
publisher = "International Statistical Institute",
number = "1",

}

TY - JOUR

T1 - Multiple testing of local maxima for detection of peaks on the (celestial) sphere

AU - Cheng, Dan

AU - Cammarota, Valentina

AU - Fantaye, Yabebal

AU - Marinucci, Domenico

AU - Schwartzman, Armin

PY - 2020/1/1

Y1 - 2020/1/1

N2 - We present a topological multiple testing scheme for detecting peaks on the sphere under isotropic Gaussian noise, where tests are performed at local maxima of the observed field filtered by the spherical needlet transform. Our setting is different from the standard Euclidean large domain asymptotic framework, yet highly relevant to realistic experimental circumstances for some important areas of application in astronomy, namely point-source detection in cosmic Microwave Background radiation (CMB) data. Motivated by this application, we shall focus on cases where a single realization of a smooth isotropic Gaussian random field on the sphere is observed, and a number of well-localized signals are superimposed on such background field. The proposed algorithms, combined with the Benjamini-Hochberg procedure for thresholding p-values, provide asymptotic control of the False Discovery Rate (FDR) and power consistency as the signal strength and the frequency of the needlet transform get large.

AB - We present a topological multiple testing scheme for detecting peaks on the sphere under isotropic Gaussian noise, where tests are performed at local maxima of the observed field filtered by the spherical needlet transform. Our setting is different from the standard Euclidean large domain asymptotic framework, yet highly relevant to realistic experimental circumstances for some important areas of application in astronomy, namely point-source detection in cosmic Microwave Background radiation (CMB) data. Motivated by this application, we shall focus on cases where a single realization of a smooth isotropic Gaussian random field on the sphere is observed, and a number of well-localized signals are superimposed on such background field. The proposed algorithms, combined with the Benjamini-Hochberg procedure for thresholding p-values, provide asymptotic control of the False Discovery Rate (FDR) and power consistency as the signal strength and the frequency of the needlet transform get large.

KW - CMB

KW - False discovery rate

KW - Gaussian random fields

KW - Height distribution

KW - Needlet transform

KW - Overshoot distribution

KW - Power

KW - Sphere

UR - http://www.scopus.com/inward/record.url?scp=85076569893&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85076569893&partnerID=8YFLogxK

U2 - 10.3150/18-BEJ1068

DO - 10.3150/18-BEJ1068

M3 - Article

AN - SCOPUS:85076569893

VL - 26

SP - 31

EP - 60

JO - Bernoulli

JF - Bernoulli

SN - 1350-7265

IS - 1

ER -