Confidence Interval Estimation of Proportion Near Zero or One: A Modeling Secondary Study
Background and Objectives: When computing a confidence interval for a binomial proportion p, one must choose an exact interval that has a coverage probability of at least 1-α for all values of p. In this study, we compared the confidence intervals of Clopper-Pearson, Wald, Wilson, and double ArcSin...
Main Authors: | , , |
---|---|
Format: | Article |
Language: | fas |
Published: |
Tehran University of Medical Sciences
2017-02-01
|
Series: | مجله اپیدمیولوژی ایران |
Subjects: | |
Online Access: | http://irje.tums.ac.ir/article-1-5621-en.html |
_version_ | 1818394864107126784 |
---|---|
author | S Hamzeh AR Soltanian J Faradmal |
author_facet | S Hamzeh AR Soltanian J Faradmal |
author_sort | S Hamzeh |
collection | DOAJ |
description | Background and Objectives: When computing a confidence interval for a binomial proportion p, one must choose an exact interval that has a coverage probability of at least 1-α for all values of p. In this study, we compared the confidence intervals of Clopper-Pearson, Wald, Wilson, and double ArcSin transformation in terms of maintaining a constant nominal type I error.
Methods: Simulations were used to compare four methods of estimating a confidence interval, including the Clopper-Pearson, Wald, Wilson, and double ArcSic. The data were generated from the binomial and Poison distribution with parameters p, n and µ=np, 1000 were produced . Type I error of each method was calculated per simulation. The above methods were used to estimate confidence intervals in a meta-analysis study.
Results: The results of the simulation study showed that double ArcSin keep confidence interval at [0,1], but for some proportion has high type I error or low coverage probability. The Clopper–Pearson interval guarantees that the coverage probability is always equal to or above the nominal confidence level for any fixed p.
Conclusion: This study showed that confidence interval estimations the Clopper-Pearson than other methods of calculating the type I error fixed and smaller. |
first_indexed | 2024-12-14T06:07:58Z |
format | Article |
id | doaj.art-65ee0448b5b34b60a65c4883e80898cb |
institution | Directory Open Access Journal |
issn | 1735-7489 2228-7507 |
language | fas |
last_indexed | 2024-12-14T06:07:58Z |
publishDate | 2017-02-01 |
publisher | Tehran University of Medical Sciences |
record_format | Article |
series | مجله اپیدمیولوژی ایران |
spelling | doaj.art-65ee0448b5b34b60a65c4883e80898cb2022-12-21T23:14:14ZfasTehran University of Medical Sciencesمجله اپیدمیولوژی ایران1735-74892228-75072017-02-011245563Confidence Interval Estimation of Proportion Near Zero or One: A Modeling Secondary StudyS Hamzeh0AR Soltanian1J Faradmal2 کارشناسی ارشد گروه آمار زیستی و اپیدمیولوژی، دانشکده بهداشت، دانشگاه علوم پزشکی همدان، همدان، ایران دانشیار آمار زیستی، گروه آمار زیستی، مرکز تحقیقات مدلسازی بیماریهای غیر واگیر، دانشکده بهداشت، دانشگاه علوم پزشکی همدان، همدان، ایران دانشیار آمار زیستی، گروه آمار زیستی، مرکز تحقیقات مدلسازی بیماریهای غیر واگیر، دانشکده بهداشت، دانشگاه علوم پزشکی همدان، همدان، ایران Background and Objectives: When computing a confidence interval for a binomial proportion p, one must choose an exact interval that has a coverage probability of at least 1-α for all values of p. In this study, we compared the confidence intervals of Clopper-Pearson, Wald, Wilson, and double ArcSin transformation in terms of maintaining a constant nominal type I error. Methods: Simulations were used to compare four methods of estimating a confidence interval, including the Clopper-Pearson, Wald, Wilson, and double ArcSic. The data were generated from the binomial and Poison distribution with parameters p, n and µ=np, 1000 were produced . Type I error of each method was calculated per simulation. The above methods were used to estimate confidence intervals in a meta-analysis study. Results: The results of the simulation study showed that double ArcSin keep confidence interval at [0,1], but for some proportion has high type I error or low coverage probability. The Clopper–Pearson interval guarantees that the coverage probability is always equal to or above the nominal confidence level for any fixed p. Conclusion: This study showed that confidence interval estimations the Clopper-Pearson than other methods of calculating the type I error fixed and smaller.http://irje.tums.ac.ir/article-1-5621-en.htmlbinomial distributionproportionexact confidence intervalapproximate confidence interval |
spellingShingle | S Hamzeh AR Soltanian J Faradmal Confidence Interval Estimation of Proportion Near Zero or One: A Modeling Secondary Study مجله اپیدمیولوژی ایران binomial distribution proportion exact confidence interval approximate confidence interval |
title | Confidence Interval Estimation of Proportion Near Zero or One: A Modeling Secondary Study |
title_full | Confidence Interval Estimation of Proportion Near Zero or One: A Modeling Secondary Study |
title_fullStr | Confidence Interval Estimation of Proportion Near Zero or One: A Modeling Secondary Study |
title_full_unstemmed | Confidence Interval Estimation of Proportion Near Zero or One: A Modeling Secondary Study |
title_short | Confidence Interval Estimation of Proportion Near Zero or One: A Modeling Secondary Study |
title_sort | confidence interval estimation of proportion near zero or one a modeling secondary study |
topic | binomial distribution proportion exact confidence interval approximate confidence interval |
url | http://irje.tums.ac.ir/article-1-5621-en.html |
work_keys_str_mv | AT shamzeh confidenceintervalestimationofproportionnearzerooroneamodelingsecondarystudy AT arsoltanian confidenceintervalestimationofproportionnearzerooroneamodelingsecondarystudy AT jfaradmal confidenceintervalestimationofproportionnearzerooroneamodelingsecondarystudy |