Based on the test of Wied et al. (2012), we construct a bootstrapping CUSUM test for correlation change. The bootstrap test uses the bootstrap critical value obtained from the distribution of the moving block bootstrap samples. The asymptotic null distribution of the bootstrap test is shown to be the same as that of the original test. Consistency of the bootstrap test is proved under an alternative hypothesis of a correlation change. A Monte Carlo simulation shows that the proposed bootstrap test has a good size performance while the existing tests have serious size distortion for conditionally heteroscedastic samples and for serially correlated samples. The better size of the bootstrap test than the existing tests is achieved at the cost of some power loss in some cases.
Bibliographical noteFunding Information:
The authors are very thankful for the fact that the paper improves a lot by exploiting the constructive comments of two referees. This study was supported by a grant from the National Research Foundation of Korea ( 2016R1A2B4008780 ) and Basic Science Research Program through the National Research Foundation of Korea (NRF) funded by the Ministry of Education ( 2009-0093827 ).
© 2019 Elsevier B.V.
- Bootstrap test
- CUSUM test
- Conditional heteroscedasticity
- Moving block bootstrapping
- Serially correlation