连续变量的差分私有拟合优度测试

IF 2 Q2 ECONOMICS
Seung Woo Kwak , Jeongyoun Ahn , Jaewoo Lee , Cheolwoo Park
{"title":"连续变量的差分私有拟合优度测试","authors":"Seung Woo Kwak ,&nbsp;Jeongyoun Ahn ,&nbsp;Jaewoo Lee ,&nbsp;Cheolwoo Park","doi":"10.1016/j.ecosta.2021.09.007","DOIUrl":null,"url":null,"abstract":"<div><p>Data privacy is a growing concern in modern data analyses as more and more types of information about individuals are collected and shared. Statistical analysis in consideration of privacy is thus becoming an exciting area of research. Differential privacy can provide a means by which one can measure the stochastic risk of violating the privacy of individuals that can result from conducting an analysis, such as a simple query from a database and a hypothesis test. The main interest of the work is a goodness-of-fit test that compares the sampled data to a known distribution. Many differentially private goodness-of-fit tests have been proposed for discrete random variables, but little work has been done for continuous variables. The objective is to review some existing tests that guarantee differential privacy for discrete random variables, and to propose an extension to continuous cases via a discretization process. The proposed test procedures are demonstrated through simulated examples and applied to the Household Financial Welfare Survey of South Korea in 2018.</p></div>","PeriodicalId":54125,"journal":{"name":"Econometrics and Statistics","volume":"31 ","pages":"Pages 81-99"},"PeriodicalIF":2.0000,"publicationDate":"2024-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":"{\"title\":\"Differentially Private Goodness-of-Fit Tests for Continuous Variables\",\"authors\":\"Seung Woo Kwak ,&nbsp;Jeongyoun Ahn ,&nbsp;Jaewoo Lee ,&nbsp;Cheolwoo Park\",\"doi\":\"10.1016/j.ecosta.2021.09.007\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"<div><p>Data privacy is a growing concern in modern data analyses as more and more types of information about individuals are collected and shared. Statistical analysis in consideration of privacy is thus becoming an exciting area of research. Differential privacy can provide a means by which one can measure the stochastic risk of violating the privacy of individuals that can result from conducting an analysis, such as a simple query from a database and a hypothesis test. The main interest of the work is a goodness-of-fit test that compares the sampled data to a known distribution. Many differentially private goodness-of-fit tests have been proposed for discrete random variables, but little work has been done for continuous variables. The objective is to review some existing tests that guarantee differential privacy for discrete random variables, and to propose an extension to continuous cases via a discretization process. The proposed test procedures are demonstrated through simulated examples and applied to the Household Financial Welfare Survey of South Korea in 2018.</p></div>\",\"PeriodicalId\":54125,\"journal\":{\"name\":\"Econometrics and Statistics\",\"volume\":\"31 \",\"pages\":\"Pages 81-99\"},\"PeriodicalIF\":2.0000,\"publicationDate\":\"2024-07-01\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"0\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"Econometrics and Statistics\",\"FirstCategoryId\":\"1085\",\"ListUrlMain\":\"https://www.sciencedirect.com/science/article/pii/S2452306221001143\",\"RegionNum\":0,\"RegionCategory\":null,\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"Q2\",\"JCRName\":\"ECONOMICS\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"Econometrics and Statistics","FirstCategoryId":"1085","ListUrlMain":"https://www.sciencedirect.com/science/article/pii/S2452306221001143","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q2","JCRName":"ECONOMICS","Score":null,"Total":0}
引用次数: 0

摘要

随着越来越多类型的个人信息被收集和共享,数据隐私在现代数据分析中日益受到关注。因此,考虑隐私的统计分析正成为一个令人兴奋的研究领域。差分隐私可以提供一种方法,用来衡量进行分析(如从数据库中进行简单查询和假设检验)时可能产生的侵犯个人隐私的随机风险。这项工作的主要关注点是拟合优度测试,它将采样数据与已知分布进行比较。针对离散型随机变量,已经提出了许多不同的私有拟合优度检验,但针对连续型变量的研究还很少。本论文旨在回顾一些保证离散随机变量差分隐私性的现有检验方法,并提出通过离散化过程扩展到连续变量的方法。通过模拟示例演示了所提出的测试程序,并将其应用于 2018 年韩国家庭金融福利调查。
本文章由计算机程序翻译,如有差异,请以英文原文为准。
Differentially Private Goodness-of-Fit Tests for Continuous Variables

Data privacy is a growing concern in modern data analyses as more and more types of information about individuals are collected and shared. Statistical analysis in consideration of privacy is thus becoming an exciting area of research. Differential privacy can provide a means by which one can measure the stochastic risk of violating the privacy of individuals that can result from conducting an analysis, such as a simple query from a database and a hypothesis test. The main interest of the work is a goodness-of-fit test that compares the sampled data to a known distribution. Many differentially private goodness-of-fit tests have been proposed for discrete random variables, but little work has been done for continuous variables. The objective is to review some existing tests that guarantee differential privacy for discrete random variables, and to propose an extension to continuous cases via a discretization process. The proposed test procedures are demonstrated through simulated examples and applied to the Household Financial Welfare Survey of South Korea in 2018.

求助全文
通过发布文献求助,成功后即可免费获取论文全文。 去求助
来源期刊
CiteScore
3.10
自引率
10.50%
发文量
84
期刊介绍: Econometrics and Statistics is the official journal of the networks Computational and Financial Econometrics and Computational and Methodological Statistics. It publishes research papers in all aspects of econometrics and statistics and comprises of the two sections Part A: Econometrics and Part B: Statistics.
×
引用
GB/T 7714-2015
复制
MLA
复制
APA
复制
导出至
BibTeX EndNote RefMan NoteFirst NoteExpress
×
提示
您的信息不完整,为了账户安全,请先补充。
现在去补充
×
提示
您因"违规操作"
具体请查看互助需知
我知道了
×
提示
确定
请完成安全验证×
copy
已复制链接
快去分享给好友吧!
我知道了
右上角分享
点击右上角分享
0
联系我们:info@booksci.cn Book学术提供免费学术资源搜索服务,方便国内外学者检索中英文文献。致力于提供最便捷和优质的服务体验。 Copyright © 2023 布克学术 All rights reserved.
京ICP备2023020795号-1
ghs 京公网安备 11010802042870号
Book学术文献互助
Book学术文献互助群
群 号:481959085
Book学术官方微信