Testing for Time-Varying Properties Under Misspecified Conditional Mean and Variance
Testing for Time-Varying Properties Under Misspecified Conditional Mean and Variance
复制标题
DOI：
10.1007/s10614-020-10014-4
复制
发表时间：
2019-07
影响因子：
2
通讯作者：
Daiki Maki;Y. Ota
中科院分区：
文献类型：
--
作者：
Daiki Maki;Y. Ota
本研究探讨了统计性能的测试下错误指定的条件均值和方差随时间变化的属性。当在数据没有时变均值但有时变方差的情况下检验条件均值的时变特性时，渐近检验具有尺寸失真。通过使用自举方法使失真最小化。类似地，当在数据具有时变均值但不具有时变方差的情况下测试条件方差的时变特性时，渐近测试具有大尺寸失真。即使使用自举方法，失真也没有得到改善。我们表明，测试的条件均值的时变特性的自助是强大的，无论随时间变化的方差模型，而测试的条件方差的时变特性不存在错误指定的时变均值的表现。
This study examines statistical performance of tests for time-varying properties under misspecified conditional mean and variance. When time-varying properties of the conditional mean are tested in the case in which data have no time-varying mean but have time-varying variance, asymptotic tests have size distortions. The distortions are minimized by the use of a bootstrap method. Similarly, when time-varying properties of the conditional variance are tested in the case in which data have time-varying mean but no time-varying variance, asymptotic tests have large size distortions. The distortions are not improved even by the use of bootstrap methods. We show that tests for time-varying properties of the conditional mean by the bootstrap are robust regardless of the time-varying variance model, whereas tests for time-varying properties of the conditional variance do not perform well in the presence of misspecified time-varying mean.