ASYMPTOTIC PROPERTIES OF RANDOM CENTRAL ORDER STATISTICS UNDER CONTAMINATION

  • Published : 2001.05.01

Abstract

Under contamination, Bahadur representations with a strong remainder term are derived for random central order statistics with a prescribed limiting rank, and asymptotic normalities for these statistics of truncated and contaminated data are proved, with a suitable limiting rank. From these results, an application to the fixed-width confidence interval problem is available.

Keywords

References

  1. Commun. Statist.-Theory Meth. v.22 no.11 Asymptotic properties of quantiles for truncated and contaminated data Fang Dong
  2. J. Statist. Plann. Inference v.22 Sequential fixed-width confidence intervals for quantiles in the presences of censoring I.Gijbels;N.Veraverbeke
  3. Stopped Random Walks A.Gut
  4. Journal of the Korean Statistical Society v.28 Sequential confidence intervals for quantiles based on recursive density estimators S.K.Kim;S.L.Kim
  5. IMS Lecture Notes Limit theorems for random central order statistics, Adaptive Statistical Procedures and Related Topics M.L.Puri;S.S.Ralescu
  6. Approximation Theorems of Mathematical Statistics R.J.Serfling
  7. CBMS-NSF Regional Conference Series in Applied Math Nonlinear Renewal Theory in Sequential Analysis M.Woodroofe