Uniform and bootstrap resampling of empirical distributions

Russell Richard Barton, Lee W. Schruben

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

Stochastic simulation models are used to predict the behavior of real systems whose components have random variation. The simulation model generates artificial random quantities based on the nature of the random variation in the real system. Very often, the probability distributions occurring in the real system are unknown, and must be estimated using finite samples. This paper presents two ways to estimate simulation model output errors due to the errors in the empirical distributions used to drive the simulation. These approaches are applied to simulations of the M/M/1 queue with an empirically sampled interarrivai time. They capture components of variance in the estimate of mean time in the system that are ignored when the empirical distribution is treated as the true distribution.

Original languageEnglish (US)
Title of host publicationProceedings of the 25th Conference on Winter Simulation, WSC 1993
EditorsWilliam E. Biles, Gerald W. Evans, Edward C. Russell, Mansooreh Mollaghasemi
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages503-508
Number of pages6
ISBN (Electronic)078031381X
DOIs
StatePublished - Dec 1 1993
Event25th Conference on Winter Simulation, WSC 1993 - Los Angeles, United States
Duration: Dec 12 1993Dec 15 1993

Publication series

NameProceedings - Winter Simulation Conference
VolumePart F129590
ISSN (Print)0891-7736

Other

Other25th Conference on Winter Simulation, WSC 1993
CountryUnited States
CityLos Angeles
Period12/12/9312/15/93

Fingerprint

Empirical Distribution
Resampling
Bootstrap
Simulation Model
M/M/1 Queue
Components of Variance
Probability distributions
Stochastic Simulation
Estimate
Stochastic Model
Simulation
Probability Distribution
Predict
Unknown
Output

All Science Journal Classification (ASJC) codes

  • Software
  • Modeling and Simulation
  • Computer Science Applications

Cite this

Barton, R. R., & Schruben, L. W. (1993). Uniform and bootstrap resampling of empirical distributions. In W. E. Biles, G. W. Evans, E. C. Russell, & M. Mollaghasemi (Eds.), Proceedings of the 25th Conference on Winter Simulation, WSC 1993 (pp. 503-508). (Proceedings - Winter Simulation Conference; Vol. Part F129590). Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1145/256563.256699
Barton, Russell Richard ; Schruben, Lee W. / Uniform and bootstrap resampling of empirical distributions. Proceedings of the 25th Conference on Winter Simulation, WSC 1993. editor / William E. Biles ; Gerald W. Evans ; Edward C. Russell ; Mansooreh Mollaghasemi. Institute of Electrical and Electronics Engineers Inc., 1993. pp. 503-508 (Proceedings - Winter Simulation Conference).
@inproceedings{5894b21ae33245609b86edebc73f642d,
title = "Uniform and bootstrap resampling of empirical distributions",
abstract = "Stochastic simulation models are used to predict the behavior of real systems whose components have random variation. The simulation model generates artificial random quantities based on the nature of the random variation in the real system. Very often, the probability distributions occurring in the real system are unknown, and must be estimated using finite samples. This paper presents two ways to estimate simulation model output errors due to the errors in the empirical distributions used to drive the simulation. These approaches are applied to simulations of the M/M/1 queue with an empirically sampled interarrivai time. They capture components of variance in the estimate of mean time in the system that are ignored when the empirical distribution is treated as the true distribution.",
author = "Barton, {Russell Richard} and Schruben, {Lee W.}",
year = "1993",
month = "12",
day = "1",
doi = "10.1145/256563.256699",
language = "English (US)",
series = "Proceedings - Winter Simulation Conference",
publisher = "Institute of Electrical and Electronics Engineers Inc.",
pages = "503--508",
editor = "Biles, {William E.} and Evans, {Gerald W.} and Russell, {Edward C.} and Mansooreh Mollaghasemi",
booktitle = "Proceedings of the 25th Conference on Winter Simulation, WSC 1993",
address = "United States",

}

Barton, RR & Schruben, LW 1993, Uniform and bootstrap resampling of empirical distributions. in WE Biles, GW Evans, EC Russell & M Mollaghasemi (eds), Proceedings of the 25th Conference on Winter Simulation, WSC 1993. Proceedings - Winter Simulation Conference, vol. Part F129590, Institute of Electrical and Electronics Engineers Inc., pp. 503-508, 25th Conference on Winter Simulation, WSC 1993, Los Angeles, United States, 12/12/93. https://doi.org/10.1145/256563.256699

Uniform and bootstrap resampling of empirical distributions. / Barton, Russell Richard; Schruben, Lee W.

Proceedings of the 25th Conference on Winter Simulation, WSC 1993. ed. / William E. Biles; Gerald W. Evans; Edward C. Russell; Mansooreh Mollaghasemi. Institute of Electrical and Electronics Engineers Inc., 1993. p. 503-508 (Proceedings - Winter Simulation Conference; Vol. Part F129590).

Research output: Chapter in Book/Report/Conference proceedingConference contribution

TY - GEN

T1 - Uniform and bootstrap resampling of empirical distributions

AU - Barton, Russell Richard

AU - Schruben, Lee W.

PY - 1993/12/1

Y1 - 1993/12/1

N2 - Stochastic simulation models are used to predict the behavior of real systems whose components have random variation. The simulation model generates artificial random quantities based on the nature of the random variation in the real system. Very often, the probability distributions occurring in the real system are unknown, and must be estimated using finite samples. This paper presents two ways to estimate simulation model output errors due to the errors in the empirical distributions used to drive the simulation. These approaches are applied to simulations of the M/M/1 queue with an empirically sampled interarrivai time. They capture components of variance in the estimate of mean time in the system that are ignored when the empirical distribution is treated as the true distribution.

AB - Stochastic simulation models are used to predict the behavior of real systems whose components have random variation. The simulation model generates artificial random quantities based on the nature of the random variation in the real system. Very often, the probability distributions occurring in the real system are unknown, and must be estimated using finite samples. This paper presents two ways to estimate simulation model output errors due to the errors in the empirical distributions used to drive the simulation. These approaches are applied to simulations of the M/M/1 queue with an empirically sampled interarrivai time. They capture components of variance in the estimate of mean time in the system that are ignored when the empirical distribution is treated as the true distribution.

UR - http://www.scopus.com/inward/record.url?scp=85031671570&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85031671570&partnerID=8YFLogxK

U2 - 10.1145/256563.256699

DO - 10.1145/256563.256699

M3 - Conference contribution

T3 - Proceedings - Winter Simulation Conference

SP - 503

EP - 508

BT - Proceedings of the 25th Conference on Winter Simulation, WSC 1993

A2 - Biles, William E.

A2 - Evans, Gerald W.

A2 - Russell, Edward C.

A2 - Mollaghasemi, Mansooreh

PB - Institute of Electrical and Electronics Engineers Inc.

ER -

Barton RR, Schruben LW. Uniform and bootstrap resampling of empirical distributions. In Biles WE, Evans GW, Russell EC, Mollaghasemi M, editors, Proceedings of the 25th Conference on Winter Simulation, WSC 1993. Institute of Electrical and Electronics Engineers Inc. 1993. p. 503-508. (Proceedings - Winter Simulation Conference). https://doi.org/10.1145/256563.256699