Bootstrapping methods in computing confidence interval: Real data application

Violation of standard assumptions in traditional method will create problem especially when estimating the confidence intervals. In the realm of applied statistics, bootstrapping methods have achieved widespread recognition and application. Bootstrapping is a useful approach to estimate standard err...

Full description

Bibliographic Details
Published in:AIP Conference Proceedings
Main Author: Mokhtar S.F.; Yusof Z.M.; Sapiri H.
Format: Conference paper
Language:English
Published: American Institute of Physics Inc. 2023
Online Access:https://www.scopus.com/inward/record.uri?eid=2-s2.0-85179827096&doi=10.1063%2f5.0177188&partnerID=40&md5=c619c62cbc3203d445513dd7e67d08db
id 2-s2.0-85179827096
spelling 2-s2.0-85179827096
Mokhtar S.F.; Yusof Z.M.; Sapiri H.
Bootstrapping methods in computing confidence interval: Real data application
2023
AIP Conference Proceedings
2896
1
10.1063/5.0177188
https://www.scopus.com/inward/record.uri?eid=2-s2.0-85179827096&doi=10.1063%2f5.0177188&partnerID=40&md5=c619c62cbc3203d445513dd7e67d08db
Violation of standard assumptions in traditional method will create problem especially when estimating the confidence intervals. In the realm of applied statistics, bootstrapping methods have achieved widespread recognition and application. Bootstrapping is a useful approach to estimate standard errors and also to obtain confidence intervals for location parameters. Bootstrapping is a computationally costly statistical approach that allows researchers to draw conclusions from the data without the need to fulfil the assumptions. The purpose of this study is to present the steps involved in calculating the confidence interval using the bootstrapping methods and to discuss on the findings. This paper also compares steps in traditional confidence interval, normal interval, percentile bootstrap and bootstrap-t methods by using real dataset. The result reveals that the normal interval outperformed the other methods as it produced the smallest interval length. Bootstrapping does not need assumptions about the distribution of the data. Bootstrapping is also a useful method because it relaxed from normality, independence, and constants variation assumptions. © 2023 Author(s).
American Institute of Physics Inc.
0094243X
English
Conference paper

author Mokhtar S.F.; Yusof Z.M.; Sapiri H.
spellingShingle Mokhtar S.F.; Yusof Z.M.; Sapiri H.
Bootstrapping methods in computing confidence interval: Real data application
author_facet Mokhtar S.F.; Yusof Z.M.; Sapiri H.
author_sort Mokhtar S.F.; Yusof Z.M.; Sapiri H.
title Bootstrapping methods in computing confidence interval: Real data application
title_short Bootstrapping methods in computing confidence interval: Real data application
title_full Bootstrapping methods in computing confidence interval: Real data application
title_fullStr Bootstrapping methods in computing confidence interval: Real data application
title_full_unstemmed Bootstrapping methods in computing confidence interval: Real data application
title_sort Bootstrapping methods in computing confidence interval: Real data application
publishDate 2023
container_title AIP Conference Proceedings
container_volume 2896
container_issue 1
doi_str_mv 10.1063/5.0177188
url https://www.scopus.com/inward/record.uri?eid=2-s2.0-85179827096&doi=10.1063%2f5.0177188&partnerID=40&md5=c619c62cbc3203d445513dd7e67d08db
description Violation of standard assumptions in traditional method will create problem especially when estimating the confidence intervals. In the realm of applied statistics, bootstrapping methods have achieved widespread recognition and application. Bootstrapping is a useful approach to estimate standard errors and also to obtain confidence intervals for location parameters. Bootstrapping is a computationally costly statistical approach that allows researchers to draw conclusions from the data without the need to fulfil the assumptions. The purpose of this study is to present the steps involved in calculating the confidence interval using the bootstrapping methods and to discuss on the findings. This paper also compares steps in traditional confidence interval, normal interval, percentile bootstrap and bootstrap-t methods by using real dataset. The result reveals that the normal interval outperformed the other methods as it produced the smallest interval length. Bootstrapping does not need assumptions about the distribution of the data. Bootstrapping is also a useful method because it relaxed from normality, independence, and constants variation assumptions. © 2023 Author(s).
publisher American Institute of Physics Inc.
issn 0094243X
language English
format Conference paper
accesstype
record_format scopus
collection Scopus
_version_ 1809677579469193216