Sequential Test for Change-Point in Long Memory Process

This paper proposes a modified kernel weighted variance ratio statistic to sequentially detect change-point that shifts from a stationary long memory process to a non-stationary long memory process. The limiting distribution of test statistic under the null hypothesis and its consistency under the alternative hypothesis are proved. Simulations indicate that the new method has better finite sample performance than the available method in the literature. Finally, we illustrate the new method by a set of U.S. inflation rate data.