Quantile Estimation Methods for Discrete Event Simulations of Regenerative Systems.

Abstract

D. L. Iglehart has proposed a method for estimating quantiles in regenerative processes. The method of averaging (MA) estimator is proposed and evaluated as an alternative to Iglehart's method. It is shown that the method of averaging is conceptually simpler and easier to implement than Iglehart's method, and under appropriate circumstances, the estimator has the same asymptotic normal distribution as Iglehart's estimator. Simulation results using the .9-quantile of waiting time in the M/M/1 queue show that the MA estimator is seriously biased; however, when the jackknife technique is applied, the method of averaging performs comparably with Iglehart's method. The method of averaging appears to be preferable for estimating quantiles in discrete event simulations of regenerative systems. (Author)

Document Details

Document Type
Technical Report
Publication Date
Jul 01, 1976
Accession Number
ADA028495

Entities

People

  • Andrew F. Seila

Organizations

  • University of North Carolina at Chapel Hill

Tags

DTIC Thesaurus Topics

  • Algorithms
  • Distribution Functions
  • Estimators
  • Mathematics
  • Normal Distribution
  • Simulations

Readers

  • Statistical inference.