Leo Vo~handu wrote: > > Why to worry about biases? > For last 30 years I am using bootstrapping as the really practical method > for statistical estimates. Method works with ANY statistical estimate and > ANY distribution. > If you really need variance analysis (ANOVA) only then read about degrees > of freedom and F-distribution. > Bootstrapping discovers often some hidden structure of your data you never > knew that it did exist at all. So it is really useful.
I agree that resampling techniques like bootstrapping or the jackknife may yield new insights as compared with traditional techniques. Since we are discussing the denominator in sample standard deviation, I have been writing in the context of ANOVAs. Best wishes, John ---------------------------------------------------------------------- For information about J forums see http://www.jsoftware.com/forums.htm
