Assessing the reliability of statistical software: Part II

被引：66

作者：

McCullough, BD ^{[1
]}

机构：

[1] Fed Commun Commiss, Washington, DC 20554 USA

来源：

AMERICAN STATISTICIAN | 1999年 / 53卷 / 02期

关键词：

accuracy; benchmarks; random number generation; software testing; StRD;

D O I：

10.2307/2685736

中图分类号：

O21 [概率论与数理统计]; C8 [统计学];

学科分类号：

020208 ; 070103 ; 0714 ;

摘要：

Part I outlined a methodology for assessing the reliability of three areas: estimation, random number generation, and calculation of statistical distributions. The present article applies this methodology to SAS, SPSS, and S-Plus, with attention to implementation details. Weaknesses are identified in all the random number generators, the S-Plus correlation procedure, and in the one-way ANOVA and nonlinear least squares routines of SAS and SPSS.

引用

页码：149 / 159

页数：11

共 29 条

[11]

HUNKA S, 1997, ANOVA NB

[12] On the accuracy of statistical distributions in Microsoft Excel 97 [J].