Journal of the Korean Statistical Society

The Korean Statistical Society (한국통계학회)
권호 표지

Kolmogorov-Smirnov Type Test for Change with Sample Fourier Coefficients

  • Published : 1996.03.01
Download PDF
⟨ Previous Next ⟩

Abstract

The problerm of testing for a constant mean is considered. A Kolmogorov-Smirnov type test using the sample Fourier coefficients is suggested and its asymptotic distribution is derived. A simulation study shows that the proposed test is more powerful than the cusum type test when there is more than one change-point or there is a cyclic change.

Keywords

References

  1. Normal Approximation and Asymptotic Expansions Bhattacharya, R. J.;Ranga Rao, R.
  2. Biometrika v.78 Detecting a smooth signal: Optimality of cusum based procedures Buckley, M. J.
  3. Handbook of Statistics v.7 Nonparametric methods for changepoint problems Csorgo, M;Horath, L.
  4. Journal of the Royal Statistical Society, B v.34 Components of Cramer-von Mises statistics Ⅰ Durbin, J.;Knott, M.
  5. Annals of Statistics v.20 Testing goodness-of-fit in regression via order selection criteria Eubank, R.;Hart, J. D.
  6. Biometrika v.41 Continuous inspection schemes Page, E. S.
  7. Biometrika v.44 A test for a change in a parameter occurring at an unknown point Page, E. S.
  8. Annals of Mathematical Statistics v.41 Moment inequalities for the maximum cumulative sum Serfling, R. J.
  9. Approximation Theorems of Mathematical Statistics Serfling, R. J.
  10. Journal of Statistical Planning and Inference v.9 Nonparametric statistical procedures for the shangepoint problem Wolfe, D. A.;Schechtman, E.