Weighted-average least squares (WALS): A survey

Jan R. Magnus; Giuseppe De Luca

doi:10.1111/joes.12094

Weighted-average least squares (WALS): A survey

Jan R. Magnus^*, Giuseppe De Luca

^*Corresponding author for this work

Research output: Contribution to Journal › Article › Academic › peer-review

Abstract

Model averaging has become a popular method of estimation, following increasing evidence that model selection and estimation should be treated as one joint procedure. Weighted-average least squares (WALS) is a recent model-average approach, which takes an intermediate position between frequentist and Bayesian methods, allows a credible treatment of ignorance, and is extremely fast to compute. We review the theory of WALS and discuss extensions and applications.

Original language	English
Pages (from-to)	117-148
Number of pages	32
Journal	Journal of Economic Surveys
Volume	30
Issue number	1
DOIs	https://doi.org/10.1111/joes.12094
Publication status	Published - 1 Feb 2016

Keywords

Computing time
Frequentist versus Bayesian
Least squares
Model averaging
Priors

Access to Document

10.1111/joes.12094

Persistent URL (handle)

Persistent link

Cite this

@article{32233d99351640bba41cf026d7bd9f50,

title = "Weighted-average least squares (WALS): A survey",

abstract = "Model averaging has become a popular method of estimation, following increasing evidence that model selection and estimation should be treated as one joint procedure. Weighted-average least squares (WALS) is a recent model-average approach, which takes an intermediate position between frequentist and Bayesian methods, allows a credible treatment of ignorance, and is extremely fast to compute. We review the theory of WALS and discuss extensions and applications.",

keywords = "Computing time, Frequentist versus Bayesian, Least squares, Model averaging, Priors",

author = "Magnus, \{Jan R.\} and \{De Luca\}, Giuseppe",

year = "2016",

month = feb,

day = "1",

doi = "10.1111/joes.12094",

language = "English",

volume = "30",

pages = "117--148",

journal = "Journal of Economic Surveys",

issn = "0950-0804",

publisher = "Wiley-Blackwell Publishing Ltd",

number = "1",

}

TY - JOUR

T1 - Weighted-average least squares (WALS)

T2 - A survey

AU - Magnus, Jan R.

AU - De Luca, Giuseppe

PY - 2016/2/1

Y1 - 2016/2/1

N2 - Model averaging has become a popular method of estimation, following increasing evidence that model selection and estimation should be treated as one joint procedure. Weighted-average least squares (WALS) is a recent model-average approach, which takes an intermediate position between frequentist and Bayesian methods, allows a credible treatment of ignorance, and is extremely fast to compute. We review the theory of WALS and discuss extensions and applications.

AB - Model averaging has become a popular method of estimation, following increasing evidence that model selection and estimation should be treated as one joint procedure. Weighted-average least squares (WALS) is a recent model-average approach, which takes an intermediate position between frequentist and Bayesian methods, allows a credible treatment of ignorance, and is extremely fast to compute. We review the theory of WALS and discuss extensions and applications.

KW - Computing time

KW - Frequentist versus Bayesian

KW - Least squares

KW - Model averaging

KW - Priors

UR - https://www.scopus.com/pages/publications/84953230318

UR - https://www.scopus.com/inward/citedby.url?scp=84953230318&partnerID=8YFLogxK

U2 - 10.1111/joes.12094

DO - 10.1111/joes.12094

M3 - Article

AN - SCOPUS:84953230318

SN - 0950-0804

VL - 30

SP - 117

EP - 148

JO - Journal of Economic Surveys

JF - Journal of Economic Surveys

IS - 1

ER -

Weighted-average least squares (WALS): A survey

Abstract

Keywords

Access to Document

Persistent URL (handle)

Other files and links

Fingerprint

Cite this