以数据驱动的带宽进行规格测试的最佳最小最大速率

IF 0.8 4区经济学 Q3 ECONOMICS

Econometric Reviews Pub Date : 2023-06-07 DOI:10.1080/07474938.2023.2198929

K. Hitomi, Masamune Iwasawa, Y. Nishiyama

{"title":"以数据驱动的带宽进行规格测试的最佳最小最大速率","authors":"K. Hitomi, Masamune Iwasawa, Y. Nishiyama","doi":"10.1080/07474938.2023.2198929","DOIUrl":null,"url":null,"abstract":"Abstract This study investigates optimal minimax rates of specification testing for linear and non-linear instrumental variable regression models. The test constructed by non-parametric kernel techniques can be rate optimal when bandwidths are selected appropriately. Since bandwidths are often selected in a data-dependent way in empirical studies, the rate-optimality of the test with data-driven bandwidths is investigated. While least squares cross-validation selects bandwidths that are optimal for estimation, it is shown not to be optimal for testing. Thus, we propose a novel bandwidth selection method for testing, the performance of which is investigated in a simulation study.","PeriodicalId":11438,"journal":{"name":"Econometric Reviews","volume":"42 1","pages":"487 - 512"},"PeriodicalIF":0.8000,"publicationDate":"2023-06-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":"{\"title\":\"Optimal minimax rates of specification testing with data-driven bandwidth\",\"authors\":\"K. Hitomi, Masamune Iwasawa, Y. Nishiyama\",\"doi\":\"10.1080/07474938.2023.2198929\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"Abstract This study investigates optimal minimax rates of specification testing for linear and non-linear instrumental variable regression models. The test constructed by non-parametric kernel techniques can be rate optimal when bandwidths are selected appropriately. Since bandwidths are often selected in a data-dependent way in empirical studies, the rate-optimality of the test with data-driven bandwidths is investigated. While least squares cross-validation selects bandwidths that are optimal for estimation, it is shown not to be optimal for testing. Thus, we propose a novel bandwidth selection method for testing, the performance of which is investigated in a simulation study.\",\"PeriodicalId\":11438,\"journal\":{\"name\":\"Econometric Reviews\",\"volume\":\"42 1\",\"pages\":\"487 - 512\"},\"PeriodicalIF\":0.8000,\"publicationDate\":\"2023-06-07\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"0\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"Econometric Reviews\",\"FirstCategoryId\":\"96\",\"ListUrlMain\":\"https://doi.org/10.1080/07474938.2023.2198929\",\"RegionNum\":4,\"RegionCategory\":\"经济学\",\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"Q3\",\"JCRName\":\"ECONOMICS\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"Econometric Reviews","FirstCategoryId":"96","ListUrlMain":"https://doi.org/10.1080/07474938.2023.2198929","RegionNum":4,"RegionCategory":"经济学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q3","JCRName":"ECONOMICS","Score":null,"Total":0}

引用次数: 0

摘要

摘要本文研究了线性和非线性工具变量回归模型的规格检验的最优极小极大率。当适当选择带宽时，由非参数核技术构建的测试可以是速率最优的。由于在实证研究中，带宽通常是以数据相关的方式选择的，因此研究了数据驱动带宽测试的速率最优性。虽然最小二乘交叉验证选择了最适合估计的带宽，但它被证明不是最适合测试的。因此，我们提出了一种新的测试带宽选择方法，并在仿真研究中对其性能进行了研究。

本文章由计算机程序翻译，如有差异，请以英文原文为准。

查看原文本刊更多论文

Optimal minimax rates of specification testing with data-driven bandwidth

Abstract This study investigates optimal minimax rates of specification testing for linear and non-linear instrumental variable regression models. The test constructed by non-parametric kernel techniques can be rate optimal when bandwidths are selected appropriately. Since bandwidths are often selected in a data-dependent way in empirical studies, the rate-optimality of the test with data-driven bandwidths is investigated. While least squares cross-validation selects bandwidths that are optimal for estimation, it is shown not to be optimal for testing. Thus, we propose a novel bandwidth selection method for testing, the performance of which is investigated in a simulation study.

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

Econometric Reviews 管理科学-数学跨学科应用

CiteScore

1.70

自引率

0.00%

发文量

审稿时长

>12 weeks

期刊介绍： Econometric Reviews is widely regarded as one of the top 5 core journals in econometrics. It probes the limits of econometric knowledge, featuring regular, state-of-the-art single blind refereed articles and book reviews. ER has been consistently the leader and innovator in its acclaimed retrospective and critical surveys and interchanges on current or developing topics. Special issues of the journal are developed by a world-renowned editorial board. These bring together leading experts from econometrics and beyond. Reviews of books and software are also within the scope of the journal. Its content is expressly intended to reach beyond econometrics and advanced empirical economics, to statistics and other social sciences.