Frequency coverage properties of a uniform shrinkage prior distribution

Research output: Contribution to journalArticle

2 Citations (Scopus)

Abstract

A uniform shrinkage prior (USP) distribution on the unknown variance component of a random-effects model is known to produce good frequency properties. The USP has a parameter that determines the shape of its density function, but it has been neglected whether the USP can maintain such good frequency properties regardless of the choice for the shape parameter. We investigate which choice for the shape parameter of the USP produces Bayesian interval estimates of random effects that meet their nominal confidence levels better than several existent choices in the literature. Using univariate and multivariate Gaussian hierarchical models, we show that the USP can achieve its best frequency properties when its shape parameter makes the USP behave similarly to an improper flat prior distribution on the unknown variance component.

Original languageEnglish (US)
Pages (from-to)2929-2939
Number of pages11
JournalJournal of Statistical Computation and Simulation
Volume87
Issue number15
DOIs
StatePublished - Oct 13 2017

Fingerprint

Shrinkage
Prior distribution
Coverage
Shape Parameter
Components of Variance
Unknown
Variance Components
Random Effects Model
Confidence Level
Gaussian Model
Hierarchical Model
Random Effects
Density Function
Probability density function
Univariate
Categorical or nominal
Interval
Estimate

All Science Journal Classification (ASJC) codes

  • Statistics and Probability
  • Modeling and Simulation
  • Statistics, Probability and Uncertainty
  • Applied Mathematics

Cite this

@article{399b18599d3944a1a4d45662352aac49,
title = "Frequency coverage properties of a uniform shrinkage prior distribution",
abstract = "A uniform shrinkage prior (USP) distribution on the unknown variance component of a random-effects model is known to produce good frequency properties. The USP has a parameter that determines the shape of its density function, but it has been neglected whether the USP can maintain such good frequency properties regardless of the choice for the shape parameter. We investigate which choice for the shape parameter of the USP produces Bayesian interval estimates of random effects that meet their nominal confidence levels better than several existent choices in the literature. Using univariate and multivariate Gaussian hierarchical models, we show that the USP can achieve its best frequency properties when its shape parameter makes the USP behave similarly to an improper flat prior distribution on the unknown variance component.",
author = "H. Tak",
year = "2017",
month = "10",
day = "13",
doi = "10.1080/00949655.2017.1349769",
language = "English (US)",
volume = "87",
pages = "2929--2939",
journal = "Journal of Statistical Computation and Simulation",
issn = "0094-9655",
publisher = "Taylor and Francis Ltd.",
number = "15",

}

Frequency coverage properties of a uniform shrinkage prior distribution. / Tak, H.

In: Journal of Statistical Computation and Simulation, Vol. 87, No. 15, 13.10.2017, p. 2929-2939.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Frequency coverage properties of a uniform shrinkage prior distribution

AU - Tak, H.

PY - 2017/10/13

Y1 - 2017/10/13

N2 - A uniform shrinkage prior (USP) distribution on the unknown variance component of a random-effects model is known to produce good frequency properties. The USP has a parameter that determines the shape of its density function, but it has been neglected whether the USP can maintain such good frequency properties regardless of the choice for the shape parameter. We investigate which choice for the shape parameter of the USP produces Bayesian interval estimates of random effects that meet their nominal confidence levels better than several existent choices in the literature. Using univariate and multivariate Gaussian hierarchical models, we show that the USP can achieve its best frequency properties when its shape parameter makes the USP behave similarly to an improper flat prior distribution on the unknown variance component.

AB - A uniform shrinkage prior (USP) distribution on the unknown variance component of a random-effects model is known to produce good frequency properties. The USP has a parameter that determines the shape of its density function, but it has been neglected whether the USP can maintain such good frequency properties regardless of the choice for the shape parameter. We investigate which choice for the shape parameter of the USP produces Bayesian interval estimates of random effects that meet their nominal confidence levels better than several existent choices in the literature. Using univariate and multivariate Gaussian hierarchical models, we show that the USP can achieve its best frequency properties when its shape parameter makes the USP behave similarly to an improper flat prior distribution on the unknown variance component.

UR - http://www.scopus.com/inward/record.url?scp=85022100759&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85022100759&partnerID=8YFLogxK

U2 - 10.1080/00949655.2017.1349769

DO - 10.1080/00949655.2017.1349769

M3 - Article

AN - SCOPUS:85022100759

VL - 87

SP - 2929

EP - 2939

JO - Journal of Statistical Computation and Simulation

JF - Journal of Statistical Computation and Simulation

SN - 0094-9655

IS - 15

ER -