On Tue, 01 Jul 2003 10:24:33 -0400, Bruce Weaver <[EMAIL PROTECTED]> wrote:
> Hi group, > I found this defintion of robustness in an old usenet post: > > Robustness is "the resilience of your results, in the face > of violations of your underlying assumptions." > > Does anyone know the original source for this defintion? Google.groups found that line in September 9, 1995. There were some interesting posts with "robust" and "assumptions." In November 2, 1995 (and later times), Herman Rubin posted - "My definition of robustness, modified from the original one of Box to allow risk function considerations, is " The robustness of a procedure is the degree to which its properties do not depend on assumptions which you do not wish to make." In a few minutes of looking in groups and on the web, I did not find the original by Box, but that would be one place to look. http://www.pitt.edu/~wpilib/index.html "Taxes are the price we pay for civilization." Justice Holmes. . . ================================================================= Instructions for joining and leaving this list, remarks about the problem of INAPPROPRIATE MESSAGES, and archives are available at: . http://jse.stat.ncsu.edu/ . =================================================================
