Non-area-specific adjustment factor for second-order efficient empirical Bayes confidence interval

Research output: Contribution to journalReview article

1 Citation (Scopus)

Abstract

An empirical Bayes confidence interval has high user demand in many applications. In particular, the second-order empirical Bayes confidence interval, the coverage error of which is of the third order for a large number of areas, m, is widely used in small area estimation when the sample size within each area is not large enough to make reliable direct estimates according to a design-based approach. Yoshimori and Lahiri (2014a) proposed a new type of confidence interval, called the second-order efficient empirical Bayes confidence interval, with a length less than that of the direct confidence estimated according to the design-based approach. However, this interval still has some disadvantages: (i) it is hard to use when at least one leverage value is high; (ii) many iterations tend to be required to obtain the estimators of one global model variance parameter as the number of areas, m, increases, due to the area-specific adjustment factor. To prevent such issues, this study proposes a more efficient confidence interval to allow for high leverage and reduce the number of iterations for large m. To achieve this purpose, we theoretically obtained a non-area-specific adjustment factor and the measure of uncertainty of the empirical Bayes estimator, which consist of empirical Bayes confidence interval, maintaining the existing desired properties. Moreover, we present three simulation results and real data analysis to show overall superiority of our confidence interval method over the other methods, including the one proposed in Yoshimori and Lahiri (2014a).

Original languageEnglish
Pages (from-to)67-78
Number of pages12
JournalComputational Statistics and Data Analysis
Volume116
DOIs
Publication statusPublished - Dec 2017

Fingerprint

Empirical Bayes
Confidence interval
Adjustment
Leverage
Empirical Bayes Estimator
Small Area Estimation
Iteration
Interval Methods
Confidence
Data analysis
Sample Size
Coverage
Tend
Estimator
Uncertainty
Interval
Estimate

All Science Journal Classification (ASJC) codes

  • Statistics and Probability
  • Computational Mathematics
  • Computational Theory and Mathematics
  • Applied Mathematics

Cite this

@article{c6a9225a7ce94d889118d65d297c1d80,
title = "Non-area-specific adjustment factor for second-order efficient empirical Bayes confidence interval",
abstract = "An empirical Bayes confidence interval has high user demand in many applications. In particular, the second-order empirical Bayes confidence interval, the coverage error of which is of the third order for a large number of areas, m, is widely used in small area estimation when the sample size within each area is not large enough to make reliable direct estimates according to a design-based approach. Yoshimori and Lahiri (2014a) proposed a new type of confidence interval, called the second-order efficient empirical Bayes confidence interval, with a length less than that of the direct confidence estimated according to the design-based approach. However, this interval still has some disadvantages: (i) it is hard to use when at least one leverage value is high; (ii) many iterations tend to be required to obtain the estimators of one global model variance parameter as the number of areas, m, increases, due to the area-specific adjustment factor. To prevent such issues, this study proposes a more efficient confidence interval to allow for high leverage and reduce the number of iterations for large m. To achieve this purpose, we theoretically obtained a non-area-specific adjustment factor and the measure of uncertainty of the empirical Bayes estimator, which consist of empirical Bayes confidence interval, maintaining the existing desired properties. Moreover, we present three simulation results and real data analysis to show overall superiority of our confidence interval method over the other methods, including the one proposed in Yoshimori and Lahiri (2014a).",
author = "Hirose, {Masayo Yoshimori}",
year = "2017",
month = "12",
doi = "10.1016/j.csda.2017.07.002",
language = "English",
volume = "116",
pages = "67--78",
journal = "Computational Statistics and Data Analysis",
issn = "0167-9473",
publisher = "Elsevier",

}

TY - JOUR

T1 - Non-area-specific adjustment factor for second-order efficient empirical Bayes confidence interval

AU - Hirose, Masayo Yoshimori

PY - 2017/12

Y1 - 2017/12

N2 - An empirical Bayes confidence interval has high user demand in many applications. In particular, the second-order empirical Bayes confidence interval, the coverage error of which is of the third order for a large number of areas, m, is widely used in small area estimation when the sample size within each area is not large enough to make reliable direct estimates according to a design-based approach. Yoshimori and Lahiri (2014a) proposed a new type of confidence interval, called the second-order efficient empirical Bayes confidence interval, with a length less than that of the direct confidence estimated according to the design-based approach. However, this interval still has some disadvantages: (i) it is hard to use when at least one leverage value is high; (ii) many iterations tend to be required to obtain the estimators of one global model variance parameter as the number of areas, m, increases, due to the area-specific adjustment factor. To prevent such issues, this study proposes a more efficient confidence interval to allow for high leverage and reduce the number of iterations for large m. To achieve this purpose, we theoretically obtained a non-area-specific adjustment factor and the measure of uncertainty of the empirical Bayes estimator, which consist of empirical Bayes confidence interval, maintaining the existing desired properties. Moreover, we present three simulation results and real data analysis to show overall superiority of our confidence interval method over the other methods, including the one proposed in Yoshimori and Lahiri (2014a).

AB - An empirical Bayes confidence interval has high user demand in many applications. In particular, the second-order empirical Bayes confidence interval, the coverage error of which is of the third order for a large number of areas, m, is widely used in small area estimation when the sample size within each area is not large enough to make reliable direct estimates according to a design-based approach. Yoshimori and Lahiri (2014a) proposed a new type of confidence interval, called the second-order efficient empirical Bayes confidence interval, with a length less than that of the direct confidence estimated according to the design-based approach. However, this interval still has some disadvantages: (i) it is hard to use when at least one leverage value is high; (ii) many iterations tend to be required to obtain the estimators of one global model variance parameter as the number of areas, m, increases, due to the area-specific adjustment factor. To prevent such issues, this study proposes a more efficient confidence interval to allow for high leverage and reduce the number of iterations for large m. To achieve this purpose, we theoretically obtained a non-area-specific adjustment factor and the measure of uncertainty of the empirical Bayes estimator, which consist of empirical Bayes confidence interval, maintaining the existing desired properties. Moreover, we present three simulation results and real data analysis to show overall superiority of our confidence interval method over the other methods, including the one proposed in Yoshimori and Lahiri (2014a).

UR - http://www.scopus.com/inward/record.url?scp=85027574065&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85027574065&partnerID=8YFLogxK

U2 - 10.1016/j.csda.2017.07.002

DO - 10.1016/j.csda.2017.07.002

M3 - Review article

AN - SCOPUS:85027574065

VL - 116

SP - 67

EP - 78

JO - Computational Statistics and Data Analysis

JF - Computational Statistics and Data Analysis

SN - 0167-9473

ER -