Bayesian multivariate receptor modeling software: BNFA and bayesMRM

Eun Sug Park; Eun Kyung Lee; Man Suk Oh

doi:10.1016/j.chemolab.2021.104280

Bayesian multivariate receptor modeling software: BNFA and bayesMRM

Eun Sug Park, Eun Kyung Lee, Man Suk Oh

Department of Statistics

Research output: Contribution to journal › Article › peer-review

6 Scopus citations

Abstract

We present user-friendly software tools to implement Bayesian multivariate receptor modeling in the form of a MATLAB function (BNFA) and an R package (bayesMRM). A basic model and a Markov chain Monte Carlo algorithm underlying BNFA and bayesMRM are given. An example of implementation based on real air pollution data is also provided. Users can freely choose between BNFA and bayesMRM depending on their computing platform. These tools are expected to facilitate implementation of Bayesian multivariate receptor models and/or Bayesian nonnegative factor analysis models and promote their use in chemometrics.

Original language	English
Article number	104280
Journal	Chemometrics and Intelligent Laboratory Systems
Volume	211
DOIs	https://doi.org/10.1016/j.chemolab.2021.104280
State	Published - 15 Apr 2021

Keywords

Bayesian factor analysis
JAGS
MATLAB
Multivariate receptor modeling
R
Software
Source apportionment

Access to Document

10.1016/j.chemolab.2021.104280

Cite this

@article{fff445c1848c49e6818f83bece3a709b,

title = "Bayesian multivariate receptor modeling software: BNFA and bayesMRM",

abstract = "We present user-friendly software tools to implement Bayesian multivariate receptor modeling in the form of a MATLAB function (BNFA) and an R package (bayesMRM). A basic model and a Markov chain Monte Carlo algorithm underlying BNFA and bayesMRM are given. An example of implementation based on real air pollution data is also provided. Users can freely choose between BNFA and bayesMRM depending on their computing platform. These tools are expected to facilitate implementation of Bayesian multivariate receptor models and/or Bayesian nonnegative factor analysis models and promote their use in chemometrics.",

keywords = "Bayesian factor analysis, JAGS, MATLAB, Multivariate receptor modeling, R, Software, Source apportionment",

author = "Park, {Eun Sug} and Lee, {Eun Kyung} and Oh, {Man Suk}",

note = "Funding Information: Man-Suk Oh{\textquoteright}s research was supported by the National Research Foundation of Korea (NRF) grant funded by the Korean Government ( NRF-2019R1A2C1003086 ). Eun Sug Park would like to dedicate BNFA to the memory of her long-time friend and mentor, Cliff Spiegelman, who first introduced multivariate receptor modeling and MATLAB to Eun Sug as well as inspiring her throughout collaborative research. The authors are grateful to Prof. Byron Gajewski for testing BNFA and bayesMRM. Publisher Copyright: {\textcopyright} 2021 The Author(s)",

year = "2021",

month = apr,

day = "15",

doi = "10.1016/j.chemolab.2021.104280",

language = "English",

volume = "211",

journal = "Chemometrics and Intelligent Laboratory Systems",

issn = "0169-7439",

publisher = "Elsevier",

}

TY - JOUR

T1 - Bayesian multivariate receptor modeling software

T2 - BNFA and bayesMRM

AU - Park, Eun Sug

AU - Lee, Eun Kyung

AU - Oh, Man Suk

N1 - Funding Information: Man-Suk Oh’s research was supported by the National Research Foundation of Korea (NRF) grant funded by the Korean Government ( NRF-2019R1A2C1003086 ). Eun Sug Park would like to dedicate BNFA to the memory of her long-time friend and mentor, Cliff Spiegelman, who first introduced multivariate receptor modeling and MATLAB to Eun Sug as well as inspiring her throughout collaborative research. The authors are grateful to Prof. Byron Gajewski for testing BNFA and bayesMRM. Publisher Copyright: © 2021 The Author(s)

PY - 2021/4/15

Y1 - 2021/4/15

N2 - We present user-friendly software tools to implement Bayesian multivariate receptor modeling in the form of a MATLAB function (BNFA) and an R package (bayesMRM). A basic model and a Markov chain Monte Carlo algorithm underlying BNFA and bayesMRM are given. An example of implementation based on real air pollution data is also provided. Users can freely choose between BNFA and bayesMRM depending on their computing platform. These tools are expected to facilitate implementation of Bayesian multivariate receptor models and/or Bayesian nonnegative factor analysis models and promote their use in chemometrics.

AB - We present user-friendly software tools to implement Bayesian multivariate receptor modeling in the form of a MATLAB function (BNFA) and an R package (bayesMRM). A basic model and a Markov chain Monte Carlo algorithm underlying BNFA and bayesMRM are given. An example of implementation based on real air pollution data is also provided. Users can freely choose between BNFA and bayesMRM depending on their computing platform. These tools are expected to facilitate implementation of Bayesian multivariate receptor models and/or Bayesian nonnegative factor analysis models and promote their use in chemometrics.

KW - Bayesian factor analysis

KW - JAGS

KW - MATLAB

KW - Multivariate receptor modeling

KW - R

KW - Software

KW - Source apportionment

UR - http://www.scopus.com/inward/record.url?scp=85102554064&partnerID=8YFLogxK

U2 - 10.1016/j.chemolab.2021.104280

DO - 10.1016/j.chemolab.2021.104280

M3 - Article

AN - SCOPUS:85102554064

SN - 0169-7439

VL - 211

JO - Chemometrics and Intelligent Laboratory Systems

JF - Chemometrics and Intelligent Laboratory Systems

M1 - 104280

ER -

Bayesian multivariate receptor modeling software: BNFA and bayesMRM

Abstract

Keywords

Access to Document

Fingerprint

Cite this