ok, thank you. Yves "David Moore" <[EMAIL PROTECTED]> wrote in message news:[EMAIL PROTECTED] > It is generally not recommended to use bootstrap intervals > for quantiles: because these statistics depend only on a > few observations from the data set, resampling tends to > give inaccurate results. > > I'm not an expert on resampling methods, just a user aware > of some hazards. Perhaps a real expert could give you some > help. > > David
. . ================================================================= Instructions for joining and leaving this list, remarks about the problem of INAPPROPRIATE MESSAGES, and archives are available at: . http://jse.stat.ncsu.edu/ . =================================================================
