{"title":"Fluctuation-type monitoring test for explosive behavior","authors":"Eiji Kurozumi","doi":"10.1016/j.ecosta.2023.06.007","DOIUrl":null,"url":null,"abstract":"<div><div>A fluctuation-type monitoring test for a bubble is proposed. The initial value is dealt with by either OLS or quasi-difference demeaning. The asymptotic property of the test under mildly explosive and local alternatives is investigated. It is shown that the fluctuation-type test has an advantage over the existing methods when the bubble appears mid- to late in the monitoring period or the bubble period is relatively long, whereas the CUSUM monitoring scheme performs better in view of power for an early bubble in the monitoring period. This theoretical property is supported in finite samples by Monte Carlo simulations. As none of the existing tests uniformly outperforms the others, the union of rejections strategy by combining the two or three monitoring tests is also proposed, which is shown to work well in finite samples.</div></div>","PeriodicalId":54125,"journal":{"name":"Econometrics and Statistics","volume":"37 ","pages":"Pages 230-249"},"PeriodicalIF":2.5000,"publicationDate":"2023-06-29","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/S2452306223000540","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
Abstract
A fluctuation-type monitoring test for a bubble is proposed. The initial value is dealt with by either OLS or quasi-difference demeaning. The asymptotic property of the test under mildly explosive and local alternatives is investigated. It is shown that the fluctuation-type test has an advantage over the existing methods when the bubble appears mid- to late in the monitoring period or the bubble period is relatively long, whereas the CUSUM monitoring scheme performs better in view of power for an early bubble in the monitoring period. This theoretical property is supported in finite samples by Monte Carlo simulations. As none of the existing tests uniformly outperforms the others, the union of rejections strategy by combining the two or three monitoring tests is also proposed, which is shown to work well in finite samples.
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.