Bayesian model averaging for benchmark dose estimation

Susan J. Simmons, Cuixian Chen, Xiaosong Li, Yishi Wang, Walter W Piegorsch, Qijun Fang, Bonnie Hu, G. Eddie Dunn

Research output: Contribution to journalArticle

8 Citations (Scopus)

Abstract

Benchmark dose estimation is widely used in various regulatory and industrial settings to estimate acceptable exposure levels to hazardous or toxic agents by predefining a level of excess risk (US EPA in Benchmark dose technical guidance document. Technical Report #EPA/100/R-12/001. U.S. Environmental Protection Agency, Washington, DC, 2012). Although benchmark dose estimation is a popular method for identifying exposure levels of agents, there are some limitations and cautions on use of this methodology. One such concern is choice of the underlying risk model. Recently, advances have been made using Bayesian model averaging to improve benchmark dose estimation in the face of model uncertainty. Herein we employ the strategies of Bayesian model averaging to build model averaged estimates for the benchmark dose. The methodology is demonstrated via a simulation study and with real data.

Original languageEnglish (US)
Pages (from-to)5-16
Number of pages12
JournalEnvironmental and Ecological Statistics
Volume22
Issue number1
DOIs
StatePublished - 2015

Fingerprint

Benchmark Dose
Bayesian Model Averaging
methodology
Methodology
Model Uncertainty
Estimate
Guidance
Excess
dose
Benchmark
Bayesian model averaging
Simulation Study
Model
simulation

Keywords

  • Bayesian model averaging
  • Benchmark dose estimation
  • Kernel smoothing
  • Posterior model probability

ASJC Scopus subject areas

  • Environmental Science(all)
  • Statistics, Probability and Uncertainty
  • Statistics and Probability

Cite this

Bayesian model averaging for benchmark dose estimation. / Simmons, Susan J.; Chen, Cuixian; Li, Xiaosong; Wang, Yishi; Piegorsch, Walter W; Fang, Qijun; Hu, Bonnie; Dunn, G. Eddie.

In: Environmental and Ecological Statistics, Vol. 22, No. 1, 2015, p. 5-16.

Research output: Contribution to journalArticle

Simmons, SJ, Chen, C, Li, X, Wang, Y, Piegorsch, WW, Fang, Q, Hu, B & Dunn, GE 2015, 'Bayesian model averaging for benchmark dose estimation', Environmental and Ecological Statistics, vol. 22, no. 1, pp. 5-16. https://doi.org/10.1007/s10651-014-0285-4
Simmons, Susan J. ; Chen, Cuixian ; Li, Xiaosong ; Wang, Yishi ; Piegorsch, Walter W ; Fang, Qijun ; Hu, Bonnie ; Dunn, G. Eddie. / Bayesian model averaging for benchmark dose estimation. In: Environmental and Ecological Statistics. 2015 ; Vol. 22, No. 1. pp. 5-16.
@article{73419737dbaa47c59541550b64e1e808,
title = "Bayesian model averaging for benchmark dose estimation",
abstract = "Benchmark dose estimation is widely used in various regulatory and industrial settings to estimate acceptable exposure levels to hazardous or toxic agents by predefining a level of excess risk (US EPA in Benchmark dose technical guidance document. Technical Report #EPA/100/R-12/001. U.S. Environmental Protection Agency, Washington, DC, 2012). Although benchmark dose estimation is a popular method for identifying exposure levels of agents, there are some limitations and cautions on use of this methodology. One such concern is choice of the underlying risk model. Recently, advances have been made using Bayesian model averaging to improve benchmark dose estimation in the face of model uncertainty. Herein we employ the strategies of Bayesian model averaging to build model averaged estimates for the benchmark dose. The methodology is demonstrated via a simulation study and with real data.",
keywords = "Bayesian model averaging, Benchmark dose estimation, Kernel smoothing, Posterior model probability",
author = "Simmons, {Susan J.} and Cuixian Chen and Xiaosong Li and Yishi Wang and Piegorsch, {Walter W} and Qijun Fang and Bonnie Hu and Dunn, {G. Eddie}",
year = "2015",
doi = "10.1007/s10651-014-0285-4",
language = "English (US)",
volume = "22",
pages = "5--16",
journal = "Environmental and Ecological Statistics",
issn = "1352-8505",
publisher = "Springer Netherlands",
number = "1",

}

TY - JOUR

T1 - Bayesian model averaging for benchmark dose estimation

AU - Simmons, Susan J.

AU - Chen, Cuixian

AU - Li, Xiaosong

AU - Wang, Yishi

AU - Piegorsch, Walter W

AU - Fang, Qijun

AU - Hu, Bonnie

AU - Dunn, G. Eddie

PY - 2015

Y1 - 2015

N2 - Benchmark dose estimation is widely used in various regulatory and industrial settings to estimate acceptable exposure levels to hazardous or toxic agents by predefining a level of excess risk (US EPA in Benchmark dose technical guidance document. Technical Report #EPA/100/R-12/001. U.S. Environmental Protection Agency, Washington, DC, 2012). Although benchmark dose estimation is a popular method for identifying exposure levels of agents, there are some limitations and cautions on use of this methodology. One such concern is choice of the underlying risk model. Recently, advances have been made using Bayesian model averaging to improve benchmark dose estimation in the face of model uncertainty. Herein we employ the strategies of Bayesian model averaging to build model averaged estimates for the benchmark dose. The methodology is demonstrated via a simulation study and with real data.

AB - Benchmark dose estimation is widely used in various regulatory and industrial settings to estimate acceptable exposure levels to hazardous or toxic agents by predefining a level of excess risk (US EPA in Benchmark dose technical guidance document. Technical Report #EPA/100/R-12/001. U.S. Environmental Protection Agency, Washington, DC, 2012). Although benchmark dose estimation is a popular method for identifying exposure levels of agents, there are some limitations and cautions on use of this methodology. One such concern is choice of the underlying risk model. Recently, advances have been made using Bayesian model averaging to improve benchmark dose estimation in the face of model uncertainty. Herein we employ the strategies of Bayesian model averaging to build model averaged estimates for the benchmark dose. The methodology is demonstrated via a simulation study and with real data.

KW - Bayesian model averaging

KW - Benchmark dose estimation

KW - Kernel smoothing

KW - Posterior model probability

UR - http://www.scopus.com/inward/record.url?scp=84939873747&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84939873747&partnerID=8YFLogxK

U2 - 10.1007/s10651-014-0285-4

DO - 10.1007/s10651-014-0285-4

M3 - Article

AN - SCOPUS:84939873747

VL - 22

SP - 5

EP - 16

JO - Environmental and Ecological Statistics

JF - Environmental and Ecological Statistics

SN - 1352-8505

IS - 1

ER -