A confidence interval test for the detection of structural breaks

Abstract: This paper develops a simple confidence interval test around the first difference of time series data to identify a structural break over time. Simulations show the robustness of the test. We also apply the test to the popular debate on the adoption of inflation targeting in New Zealand. Using the NZ inflation data, our test fails to find a structural break around the adoption date of the inflation targeting policy. The test rather supports an early reaction to the expected policy change. As the effectiveness of inflation targeting is a controversial issue, the insight provided by our test sheds some light on the debate. Our test has a wide range of possible applications in empirical literature. [Copyright &y& Elsevier]