Abstract
Motivated by the problem of the detection of a change point in the mean structure of yield curves, we introduce several methods to test the null hypothesis that the mean structure of a time series of curves does not change. The mean structure does not refer merely to the level of the curves, but also to their range and other aspects of their shape, most prominently concavity. The performance of the tests depends on whether possible break points in the error structure, which refers to the random variability in the aspects of the curves listed above, are taken into account or not. If they are not taken into account, an existing change point in the mean structure may fail to be detected with a large probability. The paper contains a complete asymptotic theory, a simulation study and illustrative data examples, as well as details of the numerical implementation of the testing procedures. This article is protected by copyright. All rights reserved
Published Version (
Free)
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call