Perspectives on Bayesian Methods and Big Data

Greg M. Allenby; Eric T. Bradlow; Edward I. George; John Liechty; Robert E. McCulloch

doi:10.1007/S40547-014-0017-9

Perspectives on Bayesian Methods and Big Data

Greg M. Allenby, Eric T. Bradlow, Edward I. George, John Liechty, Robert E. McCulloch

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

14 Scopus citations

Abstract

Researchers and practitioners are facing a world with ever-increasing amounts of data and analytic tools, such as Bayesian inference algorithms, must be improved to keep pace with technology. Bayesian methods have brought sub-stantial benefits to the discipline of Marketing Analytics, but there are inherent computational challenges with scaling them to Big Data. Several strategies with specific examples using additive regression trees and variable selection are discussed. In addition, the important observation is made that there are limits to the type of questions that can be answered using most of the Big Data available today.

Original language	English (US)
Pages (from-to)	169-175
Number of pages	7
Journal	Journal of Agricultural, Biological, and Environmental Statistics
Volume	1
Issue number	3
DOIs	https://doi.org/10.1007/S40547-014-0017-9
State	Published - Sep 2014
Externally published	Yes

Keywords

Bayesian statistics
Big Data
Scalable computation

ASJC Scopus subject areas

Statistics and Probability
General Environmental Science
Agricultural and Biological Sciences (miscellaneous)
General Agricultural and Biological Sciences
Statistics, Probability and Uncertainty
Applied Mathematics

Access to Document

10.1007/S40547-014-0017-9

Cite this

@article{24c02602bfe6430f8516b26dec4c19ee,

title = "Perspectives on Bayesian Methods and Big Data",

abstract = "Researchers and practitioners are facing a world with ever-increasing amounts of data and analytic tools, such as Bayesian inference algorithms, must be improved to keep pace with technology. Bayesian methods have brought sub-stantial benefits to the discipline of Marketing Analytics, but there are inherent computational challenges with scaling them to Big Data. Several strategies with specific examples using additive regression trees and variable selection are discussed. In addition, the important observation is made that there are limits to the type of questions that can be answered using most of the Big Data available today.",

keywords = "Bayesian statistics, Big Data, Scalable computation",

author = "Allenby, {Greg M.} and Bradlow, {Eric T.} and George, {Edward I.} and John Liechty and McCulloch, {Robert E.}",

note = "Publisher Copyright: {\textcopyright} Springer Science+Business Media New York 2014.",

year = "2014",

month = sep,

doi = "10.1007/S40547-014-0017-9",

language = "English (US)",

volume = "1",

pages = "169--175",

journal = "Journal of Agricultural, Biological, and Environmental Statistics",

issn = "1085-7117",

publisher = "Springer New York",

number = "3",

}

TY - JOUR

T1 - Perspectives on Bayesian Methods and Big Data

AU - Allenby, Greg M.

AU - Bradlow, Eric T.

AU - George, Edward I.

AU - Liechty, John

AU - McCulloch, Robert E.

N1 - Publisher Copyright: © Springer Science+Business Media New York 2014.

PY - 2014/9

Y1 - 2014/9

N2 - Researchers and practitioners are facing a world with ever-increasing amounts of data and analytic tools, such as Bayesian inference algorithms, must be improved to keep pace with technology. Bayesian methods have brought sub-stantial benefits to the discipline of Marketing Analytics, but there are inherent computational challenges with scaling them to Big Data. Several strategies with specific examples using additive regression trees and variable selection are discussed. In addition, the important observation is made that there are limits to the type of questions that can be answered using most of the Big Data available today.

AB - Researchers and practitioners are facing a world with ever-increasing amounts of data and analytic tools, such as Bayesian inference algorithms, must be improved to keep pace with technology. Bayesian methods have brought sub-stantial benefits to the discipline of Marketing Analytics, but there are inherent computational challenges with scaling them to Big Data. Several strategies with specific examples using additive regression trees and variable selection are discussed. In addition, the important observation is made that there are limits to the type of questions that can be answered using most of the Big Data available today.

KW - Bayesian statistics

KW - Big Data

KW - Scalable computation

UR - http://www.scopus.com/inward/record.url?scp=85142069420&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85142069420&partnerID=8YFLogxK

U2 - 10.1007/S40547-014-0017-9

DO - 10.1007/S40547-014-0017-9

M3 - Article

AN - SCOPUS:85142069420

SN - 1085-7117

VL - 1

SP - 169

EP - 175

JO - Journal of Agricultural, Biological, and Environmental Statistics

JF - Journal of Agricultural, Biological, and Environmental Statistics

IS - 3

ER -

Perspectives on Bayesian Methods and Big Data

Abstract

Keywords

ASJC Scopus subject areas

Access to Document

Other files and links

Fingerprint

Cite this