An omnibus test for the two-sample problem using the empirical characteristic function

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.

参考文章(31)
Andrey Feuerverger, Philip McDunnough, On the Efficiency of Empirical Characteristic Function Procedures Journal of the Royal Statistical Society: Series B (Methodological). ,vol. 43, pp. 20- 27 ,(1981) , 10.1111/J.2517-6161.1981.TB01143.X
Walter Rudin, Real and complex analysis ,(1966)
CONSTANCE L. WOOD, MICHELE M. ALTAVELA, Large Sample Results for Kolmogorov-Smirnov Statistics for Discrete Distributions Biometrika. ,vol. 65, pp. 235- 239 ,(1978) , 10.1093/BIOMET/65.1.235
E. J. Burr, Small-Sample Distributions of the Two-sample Cramer-Von Mises' $W^2$ and Watson's $U^2$ Annals of Mathematical Statistics. ,vol. 35, pp. 1091- 1098 ,(1964) , 10.1214/AOMS/1177703267
A. N. PETTITT, A two-sample Anderson-Darling rank statistic Biometrika. ,vol. 63, pp. 161- 168 ,(1976) , 10.1093/BIOMET/63.1.161
P. HALL, A. H. WELSH, A test for normality based on the empirical characteristic function Biometrika. ,vol. 70, pp. 485- 489 ,(1983) , 10.1093/BIOMET/71.3.655-A
Michael B. Marcus, Weak Convergence of the Empirical Characteristic Function Annals of Probability. ,vol. 9, pp. 194- 201 ,(1981) , 10.1214/AOP/1176994461
C. R. HEATHCOTE, The integrated squared error estimation of parameters Biometrika. ,vol. 64, pp. 255- 264 ,(1977) , 10.1093/BIOMET/64.2.255