Simulation output analysis via Dynamic Batch Means

Yingchieh Yeh, Bruce Schmeiser

Research output: Contribution to journalArticlepeer-review

17 Scopus citations


This paper is focused on estimating the quality of the sample mean from a steady-state simulation experiment with consideration of computational efficiency, memory requirement, and statistical efficiency. In addition, we seek methods that do not require knowing run length a priori. We develop an algorithm of nonoverlapping batch means that is implemented in fixed memory by dynamically changing both batch size and number of batches as the simulation runs. The algorithm, denoted by DBM for Dynamic Batch Means, requires computation time similar to other batch means data-collection methods, despite its fixed memory requirement. To achieve satisfactory statistical efficiency of DBM, we propose two associated estimators, qqTBM and qqPBM, of the variance of the sample mean and investigate their statistical properties. Our study shows that the estimator qqPBM with parameter w = 1 is, as a practical matter, better than the other proposed estimators.

Original languageEnglish
Pages (from-to)637-645
Number of pages9
JournalWinter Simulation Conference Proceedings
StatePublished - 2000


Dive into the research topics of 'Simulation output analysis via Dynamic Batch Means'. Together they form a unique fingerprint.

Cite this