TOLERANCE AND CONFIDENCE LIMITS FOR CLASSES OF DISTRIBUTIONS BASED ON FAILURE RATE.

Abstract

This paper extends the validity of exponential tolerance and confidence limits, under certain restrictions, to classes of distributions based on failure rate. In particular, the usual exponential lower tolerance limit is shown to be conservative for the increasing failure rate class of distributions in the range of population coverages and confidence coefficients of practical interest. Conservative confidence limits are also obtained on the mean. (Author)

Document Details

Document Type
Technical Report
Publication Date
Jan 01, 1966
Accession Number
AD0628661

Entities

People

  • Frank Proschan
  • Richard E. Barlow

Organizations

  • Boeing

Tags

DTIC Thesaurus Topics

  • Coefficients
  • Computing-Related Activities
  • Confidence Limits
  • Data Science
  • Information Science
  • Interdisciplinary Science
  • Mathematical Analysis

Fields of Study

  • Mathematics

Readers

  • Mathematics or Statistics
  • Neural Network Machine Learning.