2002_JUSFA Single Simulation Confidence Intervals Preprint

2002_JUSFA Single Simulation Confidence Intervals Preprint

Single Simulation Confidence Intervals

This paper presents a method to determine the confidence intervals of many simulation performance measures based on a single simulation. The confidence intervals of independent variables can be calculated directly. The confidence interval of performance functions of means can be calculated using the delta method, as for example for utilizations, frequencies, and throughputs. This allows the measurement of the accuracy of the utilizations, frequencies and throughputs using only a single simulation by using the variation of the underlying means of the performance function. The presented method is highly accurate, fast, and easy to apply. In addition, the method can also be used for short simulations or rare event applications, where methods based on batch means fail. Furthermore, this method can easily be implemented into existing simulation software.

  • File Name: 2002_JUSFA-Single-Simulation-Confidence-Intervals-Preprint.pdf
  • File type: pdf
  • File size: 138.72 KB
  • Date: February 9, 2015
  • SEO Title: Single Simulation Confidence Intervals
  • SEO Description: This paper presents a method to determine the confidence intervals of many simulation performance measures based on a single simulation. The confidence intervals of independent variables can be calculated directly. The confidence interval of performance functions of means can be calculated using the delta method, as for example for utilizations, frequencies, and throughputs. This allows the measurement of the accuracy of the utilizations, frequencies and throughputs using only a single simulation by using the variation of the underlying means of the performance function. The presented method is highly accurate, fast, and easy to apply. In addition, the method can also be used for short simulations or rare event applications, where methods based on batch means fail. Furthermore, this method can easily be implemented into existing simulation software.
  • SEO Keywords:
Many but not all of the images on this blog are under a Creative Commons License. If and only if the image above is under a Creative Commons license, you can find details for the different licenses at the Creative Commons. The most common licenses are Attribution CC-BY 2.0; 2.5; 3.0; and 4.0; and Attribution Share Alike CC-BY-SA 2.0; 2.5; 3.0; and 4.0;
Cookie Consent with Real Cookie Banner