An Improved Goodness-of-Fit Statistic for Sparse Multinomials

作者: Jeffrey S. Simonoff

DOI: 10.1080/01621459.1985.10478167

关键词: Test statisticMathematicsF-testAncillary statisticSufficient statisticStatisticsLikelihood-ratio testNull distributionPearson's chi-squared testStatistic

摘要: Abstract A new goodness-of-fit statistic for sparse multinomials is proposed. It assumed that the null distribution exhibits smoothness. The test based on maximum posterior estimator probability estimates of Simonoff (1983). Computer simulations are used to estimate distribution, significance levels, and power function test. shown a great improvement over standard tests if alternative

参考文章(14)
B. A. MAGUIRE, E. S. PEARSON, A. H. A. WYNN, THE TIME INTERVALS BETWEEN INDUSTRIAL ACCIDENTS Biometrika. ,vol. 39, pp. 168- 180 ,(1952) , 10.1093/BIOMET/39.1-2.168
J. A. John, N. R. Draper, On Testing for Two Outliers or One Outlier in Two-Way Tables Technometrics. ,vol. 20, pp. 69- 78 ,(1978) , 10.1080/00401706.1978.10489618
Jeffrey S. Simonoff, A Penalty Function Approach to Smoothing Large Sparse Contingency Tables Annals of Statistics. ,vol. 11, pp. 208- 218 ,(1983) , 10.1214/AOS/1176346071
Carl Morris, CENTRAL LIMIT THEOREMS FOR MULTINOMIAL SUMS Annals of Statistics. ,vol. 3, pp. 165- 188 ,(1975) , 10.1214/AOS/1176343006
J. Arthur Greenwood, David Durand, Aids for Fitting the Gamma Distribution by Maximum Likelihood Technometrics. ,vol. 2, pp. 55- 65 ,(1960) , 10.1080/00401706.1960.10489880
I. J. GOOD, Nonparametric roughness penalties for probability densities Biometrika. ,vol. 58, pp. 255- 277 ,(1971) , 10.1038/PHYSCI229029A0
A. Hald, Bernard Friedman, Statistical Theory with Engineering Applications ,(1952)