A New Class of Strongly Consistent Variance Estimators for Steady-State Simulations.
Abstract
The principal problem associated with steady-state simulation is the estimation of the variance term in an associated central limit theorem. This paper develops several strongly consistent estimates for this term using the strong approximations available for Brownian motion. A comparison of rates of convergence is given for a variety of estimators. Keywords: confidence intervals; regenerative simulation; simulation output analysis; strongly consistent estimation.
Document Details
- Document Type
- Technical Report
- Publication Date
- Oct 01, 1986
- Accession Number
- ADA178861
Entities
People
- Donald Iglehart
- Peter W. Glynn
Organizations
- Stanford University