A0937
Title: Introduction to differencing method and the application in testing no effect
Authors: Zhijian Li - Beijing Normal University- Hong Kong Baptist University United International College (China) [presenting]
Abstract: The idea of differencing to remove nonparametric effects provides a simple way to analyze the estimation and inference for nonparametric or semiparametric regression models. The difference-based methods are popular in practice and have many related applications. A novel difference-based method is introduced for testing the hypothesis of no relationship between the dependent and independent variables. Test statistics for nonparametric regression with Gaussian and non-Gaussian random errors are constructed. Further, it is demonstrated that these tests can detect local alternatives that converge to the null hypothesis at a rate close to n to the power -1/2. Simulation results show that the new tests are more powerful than existing methods, especially when the sample size is small.