<?xml version="1.0" encoding="utf-8" standalone="yes"?>
<item>
  <id>03879918</id>
  <dt>j</dt>
  <an>0552.62030</an>
  <augroup>
    <au>Fligner, Michael A.</au>
    <au>Rust, Steven W.</au>
  </augroup>
  <ti>On the independence problem and Kendall's tau.</ti>
  <so>Commun. Stat., Theory Methods 12, 1597-1607 (1983).</so>
  <py>1983</py>
  <pu>Taylor \& Francis, Philadelphia, PA</pu>
  <lagroup>
    <la>EN</la>
  </lagroup>
  <ccgroup>
    <cc>62G10</cc>
    <cc>62H20</cc>
    <cc>62G99</cc>
  </ccgroup>
  <utgroup>
    <ut>modification of Kendall's test for independence</ut>
    <ut>test for association</ut>
    <ut>bivariate distribution</ut>
    <ut>Kendall's tau</ut>
    <ut>exactly distribution-free test</ut>
  </utgroup>
  <cigroup>
  </cigroup>
  <ligroup>
    <li>doi:10.1080/03610928308828554</li>
  </ligroup>
  <abgroup>
    <ab>A modification of Kendall's test for independence is described which allows one to test for association in a bivariate distribution as measured by Kendall's tau, a property not shared by Kendall's procedure. The proposed procedure, however, still provides an exactly distribution- free test of independence. The test procedure is inverted to obtain a confidence interval for tau which has distinct advantages over the currently employed confidence interval.</ab>
    <rv></rv>
  </abgroup>
</item>