Some Improvements of the Bootstrap over the Delta Method Probability Errors for Whittle Estimators
Mosisa Aga
Journal of Mathematics Research, 2024, vol. 16, issue 5, 10
Abstract:
The purpose of this paper is to compare the coverage probability errors of the parametric bootstrap with that of the delta method for the covariance parameters of a regression model with auto-regressive fractionally integrated moving average (ARFIMA) errors. We consider the coverage probability errors of both confidence intervals (CIs) and tests based on the the plug-in Whittle maximum likelihood (PWML) estimators. We first show that, under some sets of conditions on the regression coefficients, the spectral density function, and the parameter values, the bounds on the coverage probability errors of the two-sided delta method and parametric bootstrap confidence intervals on the plug-in Whittle likelihood estimator of the covariance parameter are shown to be $O(n^{-1})$ and $o(n^{-3/2}\ln{n})$, respectively, where n is the sample size. Next, we show that those of the one-sided parametric bootstrap confidence intervals are shown to be $O(n^{-1/2})$ and $o(n^{-1}\ln{n})$, respectively. These results show that for both one-sided and two-sided confidence intervals and tests, the bootstrap provides a significant improvement over that of the delta method.
Date: 2024
References: View references in EconPapers View complete reference list from CitEc
Citations:
Downloads: (external link)
https://ccsenet.org/journal/index.php/jmr/article/download/0/0/50837/55142 (application/pdf)
https://ccsenet.org/journal/index.php/jmr/article/view/0/50837 (text/html)
Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.
Export reference: BibTeX
RIS (EndNote, ProCite, RefMan)
HTML/Text
Persistent link: https://EconPapers.repec.org/RePEc:ibn:jmrjnl:v:16:y:2024:i:5:p:10
Access Statistics for this article
More articles in Journal of Mathematics Research from Canadian Center of Science and Education Contact information at EDIRC.
Bibliographic data for series maintained by Canadian Center of Science and Education ().