Simon, Steve, PhD said on 10/1/02 10:12 AM: >4. You have to have enough data at the extremes. We might be able to fix >this if we trim the data, but this has been shown to work only in >simulations. > >And you have to be careful what you remove. If you remove the data by >trimming the edges, that works, according to Dr. Chambers. But if you remove >the data by creating evenly spaced bins on a rectangular grid and then >selecting the first observation to fall in each bin, then that makes CR >worse, according to Dr. Chambers. > >I have very little faith in the trimming approach. Selectively removing data >values based on their extremities is asking for trouble. It will create all >sorts of artefactual problems. And it will do nothing to fix all the other >problems listed in this email.
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/ . =================================================================
