1\name{KSd} 2\alias{KSd} 3\title{Approximate Critical Values for Kolmogorov-Smirnov's D} 4\description{ 5 Computes the critical value for Kolmogorov-Smirnov's \eqn{D_n}, for 6 sample sizes \eqn{n \ge 10}{n >= 10} and confidence level 95\%. 7} 8\details{ 9 Based on tables values given in the reference below. 10 For \eqn{n\le 80}{n <= 80} uses interpolations from exact values, elsewhere 11 uses asymptotic approximation. 12} 13\usage{ 14KSd(n) 15} 16\arguments{ 17 \item{n}{the sample size, \code{n >= 10}.} 18} 19\value{ 20 The critical value for D (two-sided) for significance level 0.05 (or 21 confidence level 95\%). 22} 23\references{ 24 Peter J. Bickel and Kjell A. Doksum (1977), 25 \emph{Mathematical Statistics: Basic Ideas and Selected Topics}. 26 Holden Day. 27 Section 9.6 and table IX. 28} 29\author{Kjetil Halvorsen and Martin Maechler} 30 31\seealso{Is used from \code{\link{ecdf.ksCI}}.} 32 33\examples{ 34KSd(90) 35KSd(1:9)# now works 36 37op <- par(mfrow=c(2,1)) 38 plot(KSd, 10, 150)# nice 39 abline(v = c(75,85), col = "gray") 40 plot(KSd, 79, 81, n = 1001)# *very* tiny discontinuity at 80 41par(op) 42} 43\keyword{distribution} 44