Paul, Try hard, reall hard to understand me. What constitutes an extreme is relative. When we trim, the remaining data leaves a new extreme. The data in these new extremes is more uniform but more important, contain more values and chances to give us a factorial pattern. The important thing is that we get cases where extremes of the causes occur together. To allow the causes to become correlated, there must be the opportunity for similar extremes to combine. Filling out the corners simply allows this to happen. It does not force the more extreme values of y to be a function of correlated pairs of extreme causes, but it allows it. To not allow such a possibility is rigging data. This is what most statisticians who use correlations do. It is not what statisticians who use ANOVA do. People who use anova tend to sample uniformly. Why do you ignore this point?
Bill > I find this whole issue strange. I mean, Dr. Chambers complains that > there are problems with traditional statistical approaches because of > sparse sampling at the extremes, that normally distributed data is the > classic representation of this problem. He then offers that 'trimming' > the data is a solution, to get a more uniform distribution. But, if you > trim the tails of the data set you've gone from relatively little data in > those areas to *no* data in those areas. > > Of course, I'm sure I misunderstand and Dr. Chambers will point out the > error of my ways and set me on the path of righteousness. > > Paul > . > . > ================================================================= > Instructions for joining and leaving this list, remarks about the > problem of INAPPROPRIATE MESSAGES, and archives are available at: > . http://jse.stat.ncsu.edu/ . > ================================================================= . . ================================================================= Instructions for joining and leaving this list, remarks about the problem of INAPPROPRIATE MESSAGES, and archives are available at: . http://jse.stat.ncsu.edu/ . =================================================================
