Please use this identifier to cite or link to this item: https://hdl.handle.net/10419/284790 
Year of Publication: 
2021
Citation: 
[Journal:] Journal of Time Series Analysis [ISSN:] 1467-9892 [Volume:] 42 [Issue:] 5-6 [Publisher:] John Wiley & Sons, Ltd [Place:] Oxford, UK [Year:] 2021 [Pages:] 534-553
Publisher: 
John Wiley & Sons, Ltd, Oxford, UK
Abstract: 
In this article, maximum deviations of sample autocovariances and autocorrelations from their theoretical counterparts over an increasing set of lags are considered. The asymptotic distribution of such statistics for physically dependent stationary time series, which is of Gumbel type, only depends on second‐order properties of the underlying time series. Since the autoregressive sieve bootstrap is able to mimic the second‐order structure asymptotically correctly it is an obvious problem whether the autoregressive (AR) sieve bootstrap, which has been shown to work for a number of relevant statistics in time series analysis, asymptotically works for maximum deviations of autocovariances and autocorrelations as well. This article shows that the question can be answered positively. Moreover, potential applications including spectral density estimation and an investigation of finite sample properties of the AR‐sieve bootstrap proposal by simulation are given.
Subjects: 
Autocovariance
autoregressive sieve bootstrap
stationary process
spectral density
simultaneous confidence bands
Persistent Identifier of the first edition: 
Creative Commons License: 
cc-by-nc-nd Logo
Document Type: 
Article
Document Version: 
Published Version

Files in This Item:
File
Size





Items in EconStor are protected by copyright, with all rights reserved, unless otherwise indicated.