Simulation input analysis: collecting data and estimating parameters for input distributions

  • Authors:
  • Mike Freimer;Lee Schruben

  • Affiliations:
  • Cornell University, Ithaca, NY;University of California at Berkeley, Berkeley, CA

  • Venue:
  • Proceedings of the 34th conference on Winter simulation: exploring new frontiers
  • Year:
  • 2002

Quantified Score

Hi-index 0.00

Visualization

Abstract

An early stage of a simulation study often consists of collecting data in order to parameterize the model. This paper addresses the question of how much data to collect, and from what sources. We use designed experiments to identify important unknown parameters, taking into account the current level of information about them. We develop approaches based on two common forms of analysis of variance: a fixed effects model, and a random effects model.