On 29 Aug 2003 at 20:19, Roger Koenker wrote: > On Fri, 29 Aug 2003, kjetil brinchmann halvorsen wrote: > > > But is it worth it to modify Kolmogorov-Smirnof fot estimated > > parameters? It has very low power anyhow. If the null hypothesis is > > "exponential distributio" (which is a scale family), what about using > > the quantile transformation twice > > > > new <- qnorm(pexp(old)) > > And if old is exponential, but not standard exponential?
But the shapiro-wilk test is scale invariant, and the exponential family is a scale family, so we can standardize old first without destroying the validity of the shapiro-wilk test? Kjetil Halvorsen > > url: www.econ.uiuc.edu/~roger/my.html Roger Koenker > email [EMAIL PROTECTED] Department of Economics > vox: 217-333-4558 University of Illinois > fax: 217-244-6678 Champaign, IL 61820 > > ______________________________________________ [EMAIL PROTECTED] mailing list https://www.stat.math.ethz.ch/mailman/listinfo/r-help
