Choosing sample path length and number of sample paths when starting in steady state

DOI10.1016/0167-6377(94)90070-1zbMath0818.62075OpenAlexW2048729803MaRDI QIDQ1892103

George S. Fishman

Publication date: 8 June 1995

Published in: Operations Research Letters (Search for Journal in Brave)

Full work available at URL: https://doi.org/10.1016/0167-6377(94)90070-1

zbMATH Keywords

Monte Carlo samplingautocorrelation functionsoptimal policymean estimationstrictly stationary stochastic processrelative cost ratiospecified accuracy

Mathematics Subject Classification ID

Non-Markovian processes: estimation (62M09) Monte Carlo methods (65C05)

Related Items (5)

On the computational complexity of MCMC-based estimators in large samples ⋮ Fast perfect simulation of Vervaat perpetuities ⋮ Random Construction of Interpolating Sets for High-Dimensional Integration ⋮ Projective re-normalization for improving the behavior of a homogeneous conic linear system ⋮ Approximation algorithms for the normalizing constant of Gibbs distributions

Cites Work

• Replication splitting and variance for simulating discrete-parameter stochastic processes
• Limit theorems for the method of replication
• The Efficiency of One Long Run Versus Independent Replications in Steady-State Simulation
• Toward guidelines for designing experiments in queuing simulation
• Evaluation of startup policies in simulation experiments
• Bias Considerations in Simulation Experiments

This page was built for publication: Choosing sample path length and number of sample paths when starting in steady state