作者: T.W. Epps , Kenneth J. Singleton
DOI: 10.1080/00949658608810963
关键词:
摘要: The empirical characteristic function (CF) is the Fourier transform of sample distribution function. values its real and imaginary parts at some number t are merely means cosine sine functions data, observations being multiplied by t. Given independent samples from two populations, we develop a test for two-sample problem which based on quadratic form in differences between respective components CFs samples. power CF compares favorably with that competing omnibus tests when data continuous. In discrete case procedure also applicable quite successful; this application it appears to have no competitors.