search for




 

Change point detection in panel data using block wild bootstrap-based CUSUM statistics
Journal of the Korean Data & Information Science Society 2022;33:389-98
Published online May 31, 2022;  https://doi.org/10.7465/jkdi.2022.33.3.389
© 2022 Korean Data and Information Science Society.

Hakjun Kim1 · Taewook Lee2

12Department of Statistics, Hankuk University of Foreign Studies
Correspondence to: This research was supported by Basic Science Research Program through the National Research Foundation of Korea (NRF) funded by the Ministry of Education (NRF-2016R1D1A1B01009508) and Hankuk University of Foreign Studies Research Fund of 2022.
1 Graduate student, Department of Statistics, Hankuk University of Foreign Studies, Gyeonggi-do 17035, Korea.
2 Professor, Department of Statistics, Hankuk University of Foreign Studies, Gyeonggi-do 17035, Korea. E-mail: twlee@hufs.ac.kr
Received March 29, 2022; Revised April 19, 2022; Accepted April 23, 2022.
This is an Open Access article distributed under the terms of the Creative Commons Attribution Non-Commercial License (http://creativecommons.org/licenses/by-nc/3.0) which permits unrestricted non-commercial use, distribution, and reproduction in any medium, provided the original work is properly cited.
Abstract
CUSUM test is widely known for detecting mean changes in time series data. However, CUSUM test tends to suffer from size distortions when autocorrelation exists in time series data. In this paper, we suggest CUSUM test based on block wild bootstrap method that combines wild bootstrap and block bootstrap to detect mean changes in panel data. Through a simulation study, our block wild bootstrap-based CUSUM test outperforms Jirak (2015) in terms of type I errors in a small sample of panel data with autocorrelation. In addition, empirical analysis is performed to confirm that our block wild bootstrap-based CUSUM test can be properly applied to actual data.
Keywords : Block wild bootstrap, CUSUM, mean change, panel data.