Variance Reduction for Regenerative Simulations. I. Internal Control and Stratified Sampling for Queues
Abstract
Two methods are discussed for reducing the variance of estimators of parameters of limiting distributions of stable stochastic processes in simulations. The methods are discussed in the context of the simple GI/G/1 queue. Of the two methods one, which is called an internal control variable, gives a variance reduction which is roughly uniform over values of the parameters of the process and, in particular, works well for values of rho sub 1 the traffic intensity, close to 1.
Document Details
- Document Type
- Technical Report
- Publication Date
- Aug 01, 1976
- Accession Number
- ADA030842
Entities
People
- Donald Iglehart
- Peter A. W. Lewis