Testing Granger non-causality in expectiles

IF 0.8 4区经济学 Q3 ECONOMICS

Econometric Reviews Pub Date : 2023-09-05 DOI:10.1080/07474938.2023.2246823

T. Bouezmarni, Mohamed Doukali, A. Taamouti

{"title":"Testing Granger non-causality in expectiles","authors":"T. Bouezmarni, Mohamed Doukali, A. Taamouti","doi":"10.1080/07474938.2023.2246823","DOIUrl":null,"url":null,"abstract":"This paper aims to derive a consistent test of Granger causality at a given expectile. We also propose a sup-Wald test for jointly testing Granger causality at all expectiles that has the correct asymptotic size and power properties. Expectiles have the advantage of capturing similar information as quantiles, but they also have the merit of being much more straightforward to use than quantiles, since they are de(cid:133)ne as least squares analogue of quantiles. Studying Granger causality in expectiles is practically simpler and allows us to examine the causality at all levels of the conditional distribution. Moreover, testing Granger causality at all expectiles provides a su¢ cient condition for testing Granger causality in distribution. A Monte Carlo simulation study reveals that our tests have good (cid:133)nite-sample size and power properties for a variety of data-generating processes and di⁄erent sample sizes. Finally, we provide two empirical applications to illustrate the usefulness of the proposed tests. ABSTRACT This paper aims to derive a consistent test of Granger causality at a given expectile. We also propose a sup-Wald test for jointly testing Granger causality at all expectiles that has the correct asymptotic size and power properties. Expectiles have the advantage of capturing similar information as quantiles, but they also have the merit of being much more straightforward to use than quantiles, since they are de(cid:133)ne as least squares analogue of quantiles. Studying Granger causality in expectiles is practically simpler and allows us to examine the causality at all levels of the conditional distribution. Moreover, testing Granger causality at all expectiles provides a su¢ cient condition for testing Granger causality in distribution. A Monte Carlo simulation study reveals that our tests have good (cid:133)nite-sample size and power properties for a variety of data-generating processes and di⁄erent sample sizes. Finally, we provide two empirical applications to illustrate the usefulness of the proposed tests.","PeriodicalId":11438,"journal":{"name":"Econometric Reviews","volume":"1 1","pages":""},"PeriodicalIF":0.8000,"publicationDate":"2023-09-05","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/10.1080/07474938.2023.2246823","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

Abstract

This paper aims to derive a consistent test of Granger causality at a given expectile. We also propose a sup-Wald test for jointly testing Granger causality at all expectiles that has the correct asymptotic size and power properties. Expectiles have the advantage of capturing similar information as quantiles, but they also have the merit of being much more straightforward to use than quantiles, since they are de(cid:133)ne as least squares analogue of quantiles. Studying Granger causality in expectiles is practically simpler and allows us to examine the causality at all levels of the conditional distribution. Moreover, testing Granger causality at all expectiles provides a su¢ cient condition for testing Granger causality in distribution. A Monte Carlo simulation study reveals that our tests have good (cid:133)nite-sample size and power properties for a variety of data-generating processes and di⁄erent sample sizes. Finally, we provide two empirical applications to illustrate the usefulness of the proposed tests. ABSTRACT This paper aims to derive a consistent test of Granger causality at a given expectile. We also propose a sup-Wald test for jointly testing Granger causality at all expectiles that has the correct asymptotic size and power properties. Expectiles have the advantage of capturing similar information as quantiles, but they also have the merit of being much more straightforward to use than quantiles, since they are de(cid:133)ne as least squares analogue of quantiles. Studying Granger causality in expectiles is practically simpler and allows us to examine the causality at all levels of the conditional distribution. Moreover, testing Granger causality at all expectiles provides a su¢ cient condition for testing Granger causality in distribution. A Monte Carlo simulation study reveals that our tests have good (cid:133)nite-sample size and power properties for a variety of data-generating processes and di⁄erent sample sizes. Finally, we provide two empirical applications to illustrate the usefulness of the proposed tests.

查看原文本刊更多论文

检验质料的格兰杰非因果性

本文旨在推导在给定期望值下格兰杰因果关系的一致性检验。我们还提出了一个sup-Wald检验，用于在所有具有正确渐近大小和幂性质的期望值上联合检验Granger因果关系。期望值具有捕获与分位数相似的信息的优点，但它们也具有比分位数更易于使用的优点，因为它们是de（cid:133）ne，是分位数的最小二乘模拟。研究期望值中的格兰杰因果关系实际上更简单，并使我们能够在条件分布的各个层面上检验因果关系。此外，在所有期望值上检验格兰杰因果关系为检验分布中的格兰杰因果性提供了一个充分的条件。蒙特卡罗模拟研究表明，对于各种数据生成过程和不同的样本量，我们的测试具有良好的（cid:133）nite样本量和功率特性。最后，我们提供了两个实证应用来说明所提出的测试的有用性。摘要本文旨在推导在给定期望值下格兰杰因果关系的一致性检验。我们还提出了一个sup-Wald检验，用于在所有具有正确渐近大小和幂性质的期望值上联合检验Granger因果关系。期望值具有捕获与分位数相似的信息的优点，但它们也具有比分位数更易于使用的优点，因为它们是de（cid:133）ne，是分位数的最小二乘模拟。研究期望值中的格兰杰因果关系实际上更简单，并使我们能够在条件分布的各个层面上检验因果关系。此外，在所有期望值上检验格兰杰因果关系为检验分布中的格兰杰因果性提供了一个充分的条件。蒙特卡罗模拟研究表明，对于各种数据生成过程和不同的样本量，我们的测试具有良好的（cid:133）nite样本量和功率特性。最后，我们提供了两个实证应用来说明所提出的测试的有用性。

本文章由计算机程序翻译，如有差异，请以英文原文为准。

求助全文

约1分钟内获得全文求助全文

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.