Published January 1, 2015 | Version v1
Conference paper Open

JACKKNIFED VARIANCE ESTIMATORS FOR SIMULATION OUTPUT ANALYSIS

  • 1. Georgia Inst Technol, H Milton Stewart Sch Ind & Syst Engn, Atlanta, GA 30332 USA
  • 2. North Carolina State Univ, Edward P Fitts Dept Ind & Syst Engn, Raleigh, NC 27695 USA
  • 3. Fenxi LLC, Belmont, CA 94002 USA
  • 4. AT&T, Adv Analyt, Big Data, Atlanta, GA 30319 USA

Description

We develop new point estimators for the variance parameter of a steady-state simulation process. The estimators are based on jackknifed versions of nonoverlapping batch means, overlapping batch means, and standardized time series variance estimators. The new estimators have reduced bias and can be manipulated to reduce their variance and mean-squared error compared with their predecessors, facts which we demonstrate analytically and empirically.

Files

bib-ede484f8-0056-477a-8230-d21a96a7a83c.txt

Files (196 Bytes)

Name Size Download all
md5:3bba1b46af1877ed62aa7957fd17c241
196 Bytes Preview Download