A TEST FOR WEAK STATIONARITY IN THE SPECTRAL DOMAIN
提出一种检验时间序列弱平稳性的方法，适用于局部平稳和断点模型，其渐近分布为标准布朗桥片函数，不依赖未知参数，并通过蒙特卡洛实验和通胀数据、神经科学震颤数据验证了有效性。
We examine a test for weak stationarity against alternatives that covers both local-stationarity and break point models. A key feature of the test is that its asymptotic distribution is a functional of the standard Brownian bridge sheet in [0,1] 2 , so that it does not depend on any unknown quantity. The test has nontrivial power against local alternatives converging to the null hypothesis at a T −1/2 rate, where T is the sample size. We also examine an easy-to-implement bootstrap analogue and present the finite sample performance in a Monte Carlo experiment. Finally, we implement the methodology to assess the stability of inflation dynamics in the United States and on a set of neuroscience tremor data.