Influence of a prior distribution on traffic intensity estimation with covariates

Research output: Contribution to journalArticle

6 Citations (Scopus)

Abstract

In this paper, several methods are suggested to estimate the expected traffic intensity (ρ) in M/M/1 queues with covariates. A Monte-Carlo simulation is used to generate M/M/1 queues where the arrival (service) rates are governed by both their covariate effects and the random error which follows a lognormal distribution. An ad hoc estimator (ρ̂ADL) is derived for traffic intensity based on a lognormal prior. Its performance is compared to those of the following procedures, when the true prior is a lognormal distribution: empirical Bayes estimator obtained based on a gamma prior distribution (ρ̂EBG), model based regression estimator (ρ̂M) and data based raw estimator (ρ̂R). Results of a simulation study indicate that the performance of ρ̂ADL is reasonable; the overall performance of ρ̂EBG is not significantly different from that of ρ̂ADL and ρ̂M can replace both ρ̂ADL and ρ̂EBG, when the variability of the arrival (service) rate due to random error is relatively small.

Original languageEnglish
Pages (from-to)169-180
Number of pages12
JournalJournal of Statistical Computation and Simulation
Volume55
Issue number3
DOIs
Publication statusPublished - 1996 Jan 1

Fingerprint

Random errors
Prior distribution
M/M/1 Queue
Covariates
Log Normal Distribution
Random Error
Traffic
Empirical Bayes Estimator
Estimator
Regression Estimator
Gamma distribution
Monte Carlo Simulation
Simulation Study
Model-based
Estimate
Influence
Activities of daily living
Monte Carlo simulation
Log normal distribution
Random error

All Science Journal Classification (ASJC) codes

  • Statistics and Probability
  • Modelling and Simulation
  • Statistics, Probability and Uncertainty
  • Applied Mathematics

Cite this

@article{a41152754be44afd878c52816a488507,
title = "Influence of a prior distribution on traffic intensity estimation with covariates",
abstract = "In this paper, several methods are suggested to estimate the expected traffic intensity (ρ) in M/M/1 queues with covariates. A Monte-Carlo simulation is used to generate M/M/1 queues where the arrival (service) rates are governed by both their covariate effects and the random error which follows a lognormal distribution. An ad hoc estimator (ρ̂ADL) is derived for traffic intensity based on a lognormal prior. Its performance is compared to those of the following procedures, when the true prior is a lognormal distribution: empirical Bayes estimator obtained based on a gamma prior distribution (ρ̂EBG), model based regression estimator (ρ̂M) and data based raw estimator (ρ̂R). Results of a simulation study indicate that the performance of ρ̂ADL is reasonable; the overall performance of ρ̂EBG is not significantly different from that of ρ̂ADL and ρ̂M can replace both ρ̂ADL and ρ̂EBG, when the variability of the arrival (service) rate due to random error is relatively small.",
author = "Sohn, {So Young}",
year = "1996",
month = "1",
day = "1",
doi = "10.1080/00949659608811759",
language = "English",
volume = "55",
pages = "169--180",
journal = "Journal of Statistical Computation and Simulation",
issn = "0094-9655",
publisher = "Taylor and Francis Ltd.",
number = "3",

}

Influence of a prior distribution on traffic intensity estimation with covariates. / Sohn, So Young.

In: Journal of Statistical Computation and Simulation, Vol. 55, No. 3, 01.01.1996, p. 169-180.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Influence of a prior distribution on traffic intensity estimation with covariates

AU - Sohn, So Young

PY - 1996/1/1

Y1 - 1996/1/1

N2 - In this paper, several methods are suggested to estimate the expected traffic intensity (ρ) in M/M/1 queues with covariates. A Monte-Carlo simulation is used to generate M/M/1 queues where the arrival (service) rates are governed by both their covariate effects and the random error which follows a lognormal distribution. An ad hoc estimator (ρ̂ADL) is derived for traffic intensity based on a lognormal prior. Its performance is compared to those of the following procedures, when the true prior is a lognormal distribution: empirical Bayes estimator obtained based on a gamma prior distribution (ρ̂EBG), model based regression estimator (ρ̂M) and data based raw estimator (ρ̂R). Results of a simulation study indicate that the performance of ρ̂ADL is reasonable; the overall performance of ρ̂EBG is not significantly different from that of ρ̂ADL and ρ̂M can replace both ρ̂ADL and ρ̂EBG, when the variability of the arrival (service) rate due to random error is relatively small.

AB - In this paper, several methods are suggested to estimate the expected traffic intensity (ρ) in M/M/1 queues with covariates. A Monte-Carlo simulation is used to generate M/M/1 queues where the arrival (service) rates are governed by both their covariate effects and the random error which follows a lognormal distribution. An ad hoc estimator (ρ̂ADL) is derived for traffic intensity based on a lognormal prior. Its performance is compared to those of the following procedures, when the true prior is a lognormal distribution: empirical Bayes estimator obtained based on a gamma prior distribution (ρ̂EBG), model based regression estimator (ρ̂M) and data based raw estimator (ρ̂R). Results of a simulation study indicate that the performance of ρ̂ADL is reasonable; the overall performance of ρ̂EBG is not significantly different from that of ρ̂ADL and ρ̂M can replace both ρ̂ADL and ρ̂EBG, when the variability of the arrival (service) rate due to random error is relatively small.

UR - http://www.scopus.com/inward/record.url?scp=0030298694&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0030298694&partnerID=8YFLogxK

U2 - 10.1080/00949659608811759

DO - 10.1080/00949659608811759

M3 - Article

AN - SCOPUS:0030298694

VL - 55

SP - 169

EP - 180

JO - Journal of Statistical Computation and Simulation

JF - Journal of Statistical Computation and Simulation

SN - 0094-9655

IS - 3

ER -