Communications for Statistical Applications and Methods

The Korean Statistical Society (한국통계학회)
권호 표지

Confidence Intervals for Distribution Function

  • Choi, J.R. (Department of Mathematics, Dong-A University) ;
  • Kang, M.K. (Department of Computer Science and Statistics, Dong-Eui University) ;
  • Chu, I.S. (Department of Mathematics, Dong-A University)
  • Published : 1997.04.01
Download PDF
⟨ Previous Next ⟩

Abstract

In this note we consider confidence interval based on Kolmogorov-Smirnov statistic. In order to obtain confidence interval we need percentage points of the statistics. Bootstrap method is examined whether it is useful to determine the points. It is concluded that the method is useful for observations with many ties, whereas it gives less conserbative points for continuous distributions.

Keywords

References

  1. Ann. Math. Statist. v.23 Asymptotic theory of certain goodness of fit criteria based on stochastic processes Anderson, T. W.;Darling, D. A.
  2. J. Amer. Statist. Soc. v.84 Confidence bands for a distribution function using the bootstrap Bickel, P. J.;Krieger, A. M.
  3. Ann. statist. v.7 Bootstrap methods: another look at the jadkknife Efron, B.
  4. Handbook of statistical Tables Owen, D. B.
  5. Ann. Inst. statist. Math. v.44 Integrated squared error of kernel-type estimator of distribution function Shirahata, S.;Chu, I. S.
  6. Goodness-of-Fit Techniques (ED. D'Agostino and Stephens) Tests Based on EDF statistics Stephens, M. A.