EVOLUTION-MANAGER
Edit File: Wilcoxon.html
<!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.0 Strict//EN" "http://www.w3.org/TR/xhtml1/DTD/xhtml1-strict.dtd"><html xmlns="http://www.w3.org/1999/xhtml"><head><title>R: Distribution of the Wilcoxon Rank Sum Statistic</title> <meta http-equiv="Content-Type" content="text/html; charset=utf-8" /> <link rel="stylesheet" type="text/css" href="R.css" /> </head><body> <table width="100%" summary="page for Wilcoxon {stats}"><tr><td>Wilcoxon {stats}</td><td style="text-align: right;">R Documentation</td></tr></table> <h2>Distribution of the Wilcoxon Rank Sum Statistic</h2> <h3>Description</h3> <p>Density, distribution function, quantile function and random generation for the distribution of the Wilcoxon rank sum statistic obtained from samples with size <code>m</code> and <code>n</code>, respectively. </p> <h3>Usage</h3> <pre> dwilcox(x, m, n, log = FALSE) pwilcox(q, m, n, lower.tail = TRUE, log.p = FALSE) qwilcox(p, m, n, lower.tail = TRUE, log.p = FALSE) rwilcox(nn, m, n) </pre> <h3>Arguments</h3> <table summary="R argblock"> <tr valign="top"><td><code>x, q</code></td> <td> <p>vector of quantiles.</p> </td></tr> <tr valign="top"><td><code>p</code></td> <td> <p>vector of probabilities.</p> </td></tr> <tr valign="top"><td><code>nn</code></td> <td> <p>number of observations. If <code>length(nn) > 1</code>, the length is taken to be the number required.</p> </td></tr> <tr valign="top"><td><code>m, n</code></td> <td> <p>numbers of observations in the first and second sample, respectively. Can be vectors of positive integers.</p> </td></tr> <tr valign="top"><td><code>log, log.p</code></td> <td> <p>logical; if TRUE, probabilities p are given as log(p).</p> </td></tr> <tr valign="top"><td><code>lower.tail</code></td> <td> <p>logical; if TRUE (default), probabilities are <i>P[X ≤ x]</i>, otherwise, <i>P[X > x]</i>.</p> </td></tr> </table> <h3>Details</h3> <p>This distribution is obtained as follows. Let <code>x</code> and <code>y</code> be two random, independent samples of size <code>m</code> and <code>n</code>. Then the Wilcoxon rank sum statistic is the number of all pairs <code>(x[i], y[j])</code> for which <code>y[j]</code> is not greater than <code>x[i]</code>. This statistic takes values between <code>0</code> and <code>m * n</code>, and its mean and variance are <code>m * n / 2</code> and <code>m * n * (m + n + 1) / 12</code>, respectively. </p> <p>If any of the first three arguments are vectors, the recycling rule is used to do the calculations for all combinations of the three up to the length of the longest vector. </p> <h3>Value</h3> <p><code>dwilcox</code> gives the density, <code>pwilcox</code> gives the distribution function, <code>qwilcox</code> gives the quantile function, and <code>rwilcox</code> generates random deviates. </p> <p>The length of the result is determined by <code>nn</code> for <code>rwilcox</code>, and is the maximum of the lengths of the numerical arguments for the other functions. </p> <p>The numerical arguments other than <code>nn</code> are recycled to the length of the result. Only the first elements of the logical arguments are used. </p> <h3>Warning</h3> <p>These functions can use large amounts of memory and stack (and even crash <span style="font-family: Courier New, Courier; color: #666666;"><b>R</b></span> if the stack limit is exceeded and stack-checking is not in place) if one sample is large (several thousands or more). </p> <h3>Note</h3> <p>S-PLUS uses a different (but equivalent) definition of the Wilcoxon statistic: see <code><a href="wilcox.test.html">wilcox.test</a></code> for details. </p> <h3>Author(s)</h3> <p>Kurt Hornik</p> <h3>Source</h3> <p>These ("d","p","q") are calculated via recursion, based on <code>cwilcox(k, m, n)</code>, the number of choices with statistic <code>k</code> from samples of size <code>m</code> and <code>n</code>, which is itself calculated recursively and the results cached. Then <code>dwilcox</code> and <code>pwilcox</code> sum appropriate values of <code>cwilcox</code>, and <code>qwilcox</code> is based on inversion. </p> <p><code>rwilcox</code> generates a random permutation of ranks and evaluates the statistic. Note that it is based on the same C code as <code><a href="../../base/html/sample.html">sample</a>()</code>, and hence is determined by <code><a href="../../base/html/Random.html">.Random.seed</a></code>, notably from <code><a href="../../base/html/Random.html">RNGkind</a>(sample.kind = ..)</code> which changed with <span style="font-family: Courier New, Courier; color: #666666;"><b>R</b></span> version 3.6.0. </p> <h3>See Also</h3> <p><code><a href="wilcox.test.html">wilcox.test</a></code> to calculate the statistic from data, find p values and so on. </p> <p><a href="Distributions.html">Distributions</a> for standard distributions, including <code><a href="SignRank.html">dsignrank</a></code> for the distribution of the <em>one-sample</em> Wilcoxon signed rank statistic. </p> <h3>Examples</h3> <pre> require(graphics) x <- -1:(4*6 + 1) fx <- dwilcox(x, 4, 6) Fx <- pwilcox(x, 4, 6) layout(rbind(1,2), widths = 1, heights = c(3,2)) plot(x, fx, type = "h", col = "violet", main = "Probabilities (density) of Wilcoxon-Statist.(n=6, m=4)") plot(x, Fx, type = "s", col = "blue", main = "Distribution of Wilcoxon-Statist.(n=6, m=4)") abline(h = 0:1, col = "gray20", lty = 2) layout(1) # set back N <- 200 hist(U <- rwilcox(N, m = 4,n = 6), breaks = 0:25 - 1/2, border = "red", col = "pink", sub = paste("N =",N)) mtext("N * f(x), f() = true \"density\"", side = 3, col = "blue") lines(x, N*fx, type = "h", col = "blue", lwd = 2) points(x, N*fx, cex = 2) ## Better is a Quantile-Quantile Plot qqplot(U, qw <- qwilcox((1:N - 1/2)/N, m = 4, n = 6), main = paste("Q-Q-Plot of empirical and theoretical quantiles", "Wilcoxon Statistic, (m=4, n=6)", sep = "\n")) n <- as.numeric(names(print(tU <- table(U)))) text(n+.2, n+.5, labels = tU, col = "red") </pre> <hr /><div style="text-align: center;">[Package <em>stats</em> version 3.6.0 <a href="00Index.html">Index</a>]</div> </body></html>