False posteriors for the long-term growth determinants

Wojciech W. Charemza; Rodney Strachan; Piotr Zurawski

doi:10.1016/j.econlet.2010.08.026

False posteriors for the long-term growth determinants

Wojciech W. Charemza^*, Rodney Strachan, Piotr Zurawski

^*Corresponding author for this work

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

Abstract

This paper shows: (1) bias may exist in Bayesian methods of selecting regressors; (2) the non-zero posterior inclusion probability of irrelevant regressors might affect the choice of regressors; and (3) how to compute the probability of falsely including a variable.

Original language	English
Pages (from-to)	144-146
Number of pages	3
Journal	Economics Letters
Volume	109
Issue number	3
DOIs	https://doi.org/10.1016/j.econlet.2010.08.026
Publication status	Published - Dec 2010

Access to Document

10.1016/j.econlet.2010.08.026

Cite this

@article{0e6864558e1f4dfea9b77dad0f8e48f2,

title = "False posteriors for the long-term growth determinants",

abstract = "This paper shows: (1) bias may exist in Bayesian methods of selecting regressors; (2) the non-zero posterior inclusion probability of irrelevant regressors might affect the choice of regressors; and (3) how to compute the probability of falsely including a variable.",

keywords = "Bayesian averaging, Long-term growth, Selection bias, Selection of determinants",

author = "Charemza, {Wojciech W.} and Rodney Strachan and Piotr Zurawski",

year = "2010",

month = dec,

doi = "10.1016/j.econlet.2010.08.026",

language = "English",

volume = "109",

pages = "144--146",

journal = "Economics Letters",

issn = "0165-1765",

publisher = "Elsevier B.V.",

number = "3",

}

TY - JOUR

T1 - False posteriors for the long-term growth determinants

AU - Charemza, Wojciech W.

AU - Strachan, Rodney

AU - Zurawski, Piotr

PY - 2010/12

Y1 - 2010/12

N2 - This paper shows: (1) bias may exist in Bayesian methods of selecting regressors; (2) the non-zero posterior inclusion probability of irrelevant regressors might affect the choice of regressors; and (3) how to compute the probability of falsely including a variable.

AB - This paper shows: (1) bias may exist in Bayesian methods of selecting regressors; (2) the non-zero posterior inclusion probability of irrelevant regressors might affect the choice of regressors; and (3) how to compute the probability of falsely including a variable.

KW - Bayesian averaging

KW - Long-term growth

KW - Selection bias

KW - Selection of determinants

UR - http://www.scopus.com/inward/record.url?scp=77957765257&partnerID=8YFLogxK

U2 - 10.1016/j.econlet.2010.08.026

DO - 10.1016/j.econlet.2010.08.026

M3 - Article

SN - 0165-1765

VL - 109

SP - 144

EP - 146

JO - Economics Letters

JF - Economics Letters

IS - 3

ER -

False posteriors for the long-term growth determinants

Abstract

Access to Document

Other files and links

Fingerprint

Cite this