Charles Explorer logo
🇬🇧

On detection of changes in autoregressive time series I. Asymptotics

Publication at Faculty of Mathematics and Physics |
2007

Abstract

Various test statistics are discussed which can be used for detection changes in parameters of autoregressive time series. A limiting behavior of the test statistics is derived under the null hypothesis of no change as well as under alternatives.

These asymptotic results are compared to some corresponding bootstrap procedures. A small simulation study is conducted.