In this paper, hypothesis testing for positive first-degree and higher-degree expectation dependence is investigated. Some tests of Kolmogorov–Smirnov type are constructed, which are shown to control type I error well and to be consistent against global alternative hypothesis. Further, the tests can also detect local alternative hypotheses distinct from the null hypothesis at a rate as close to the square root of the sample size as possible, which is the fastest possible rate in hypothesis testing. A nonparametric Monte Carlo test procedure is applied to implement the new tests because both sampling and limiting null distributions are not tractable. Simulation studies and a real data analysis are carried out to illustrate the performances of the new tests.
|Original language||English (US)|
|Number of pages||19|
|Journal||Annals of the Institute of Statistical Mathematics|
|State||Published - Feb 1 2016|
All Science Journal Classification (ASJC) codes
- Statistics and Probability