key steps to evaluate statistical software

by Diego Kuonen and Armin S. A. Rohrl .

Abstract: After extensive studies and many painful experiences we came to the conclusion that there is no statistics package on the market that combines performance (in speed and memory), scalability, predictability, compatibility, user-friendliness, extensibility, intelligent agents and good presentation of data and results. We explain why we consider these points to be crucial, especially to statisticians, and compare the current most-used programs. Finally we conclude that it is time to design a new statistics package incorporating modern ideas from computer science.

Key Words: Compatibility, Distributed, Extensibility, Genstat, Intelligent agents, Minitab, Multi-Threaded, Ox, Performance, Predictability, Presentation, R, SAS, S-Plus, Scalability, SPSS, Statistics software, User-friendliness, Xlisp-Stat, XploRe

Authors:
Diego Kuonen, Diego.Kuonen@epfl.c
Armin S. A. Rohrl, Armin.Roehrl@epfl.c

Editor: John Hinde , jph@msor.exeter.ac.uk

READING THE ARTICLE: You can read the article in portable document (.pdf) format (194698 bytes.)

NOTE: The content of this article is the intellectual property of the authors, who retains all rights to future publication.

This page has been accessed 408 times since April 1, 2007.

Return to the InterStat Home Page.