Median Bounds and Their Application

Research output: Contribution to journalArticle

Abstract

Basic methods are given to evaluate or estimate the median for various probability distributions. These methods are then applied to determine the precise median of several nontrivial distributions, including weighted selection and the sum of heterogeneous Bernoulli trials conditioned to lie within any interval centered about the mean. These bounds are then used to give simple analyses of algorithms such as interpolation search and some aspects of PRAM emulation.

Original languageEnglish (US)
Pages (from-to)184-236
Number of pages53
JournalJournal of Algorithms
Volume38
Issue number1
DOIs
StatePublished - Jan 2001

    Fingerprint

Keywords

  • Bernoulli trials
  • Conditioned Bernoulli trials
  • Conditioned hypergeometric distribution
  • Hypergeometric distribution
  • Interpolation search
  • Median
  • PRAM emulation
  • Weighted selection

ASJC Scopus subject areas

  • Control and Optimization
  • Computational Mathematics
  • Computational Theory and Mathematics

Cite this