A Non-Bayesian Approach To Estimation Using Tested Priors

Varghese George, S. K. Katti

Research output: Contribution to journalArticle

Abstract

A two stage sampling scheme for estimating the mean of a distribution, proposed by Katti (1962), is investigated and some of its properties are generalized. The method utilizes subjective information in the analysis, but is still within the classical framework. The generalized mean square error of the estimator is computed as the loss function and is compared with that of the usual classical approach. Modifications are suggested to improve the efficiency of the estimator by incorporating the uncertainty of the subjective information. The procedure, which is referred to as “the Method of Tested Priors” is an alternate way of using subjective information without necessarily agreeing with the philosophical aspect of the Bayesian approach.

Original languageEnglish (US)
Pages (from-to)1629-1639
Number of pages11
JournalCommunications in Statistics - Theory and Methods
Volume17
Issue number5
DOIs
StatePublished - Jan 1 1988
Externally publishedYes

Fingerprint

Two-stage Sampling
Estimator
Loss Function
Bayesian Approach
Mean square error
Alternate
Uncertainty
Framework

Keywords

  • efficiency
  • generalized mean square error
  • initial estimate
  • non-central chi equare
  • optimal region
  • prior variance

ASJC Scopus subject areas

  • Statistics and Probability

Cite this

A Non-Bayesian Approach To Estimation Using Tested Priors. / George, Varghese; Katti, S. K.

In: Communications in Statistics - Theory and Methods, Vol. 17, No. 5, 01.01.1988, p. 1629-1639.

Research output: Contribution to journalArticle

@article{9d0a6e430c2041b292f49feb3a059bee,
title = "A Non-Bayesian Approach To Estimation Using Tested Priors",
abstract = "A two stage sampling scheme for estimating the mean of a distribution, proposed by Katti (1962), is investigated and some of its properties are generalized. The method utilizes subjective information in the analysis, but is still within the classical framework. The generalized mean square error of the estimator is computed as the loss function and is compared with that of the usual classical approach. Modifications are suggested to improve the efficiency of the estimator by incorporating the uncertainty of the subjective information. The procedure, which is referred to as “the Method of Tested Priors” is an alternate way of using subjective information without necessarily agreeing with the philosophical aspect of the Bayesian approach.",
keywords = "efficiency, generalized mean square error, initial estimate, non-central chi equare, optimal region, prior variance",
author = "Varghese George and Katti, {S. K.}",
year = "1988",
month = "1",
day = "1",
doi = "10.1080/03610928808829702",
language = "English (US)",
volume = "17",
pages = "1629--1639",
journal = "Communications in Statistics - Theory and Methods",
issn = "0361-0926",
publisher = "Taylor and Francis Ltd.",
number = "5",

}

TY - JOUR

T1 - A Non-Bayesian Approach To Estimation Using Tested Priors

AU - George, Varghese

AU - Katti, S. K.

PY - 1988/1/1

Y1 - 1988/1/1

N2 - A two stage sampling scheme for estimating the mean of a distribution, proposed by Katti (1962), is investigated and some of its properties are generalized. The method utilizes subjective information in the analysis, but is still within the classical framework. The generalized mean square error of the estimator is computed as the loss function and is compared with that of the usual classical approach. Modifications are suggested to improve the efficiency of the estimator by incorporating the uncertainty of the subjective information. The procedure, which is referred to as “the Method of Tested Priors” is an alternate way of using subjective information without necessarily agreeing with the philosophical aspect of the Bayesian approach.

AB - A two stage sampling scheme for estimating the mean of a distribution, proposed by Katti (1962), is investigated and some of its properties are generalized. The method utilizes subjective information in the analysis, but is still within the classical framework. The generalized mean square error of the estimator is computed as the loss function and is compared with that of the usual classical approach. Modifications are suggested to improve the efficiency of the estimator by incorporating the uncertainty of the subjective information. The procedure, which is referred to as “the Method of Tested Priors” is an alternate way of using subjective information without necessarily agreeing with the philosophical aspect of the Bayesian approach.

KW - efficiency

KW - generalized mean square error

KW - initial estimate

KW - non-central chi equare

KW - optimal region

KW - prior variance

UR - http://www.scopus.com/inward/record.url?scp=84948496247&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84948496247&partnerID=8YFLogxK

U2 - 10.1080/03610928808829702

DO - 10.1080/03610928808829702

M3 - Article

VL - 17

SP - 1629

EP - 1639

JO - Communications in Statistics - Theory and Methods

JF - Communications in Statistics - Theory and Methods

SN - 0361-0926

IS - 5

ER -