The deviance information criterion in comparison of normal mixing models

Thomas Fung; Joanna J. J. Wang; Eugene Seneta

doi:10.1111/insr.12063

The deviance information criterion in comparison of normal mixing models

Thomas Fung, Joanna J. J. Wang, Eugene Seneta

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

1 Citation (Scopus)

Abstract

Model selection from several non-nested models by using the deviance information criterion within Bayesian inference Using Gibbs Sampling (BUGS) software needs to be treated with caution. This is particularly important if one can specify a model in various mixing representations, as for the normal variance-mean mixing distribution occurring in financial contexts. We propose a procedure to compare goodness of fit of several non-nested models, which uses BUGS software in part.

Original language	English
Pages (from-to)	411-421
Number of pages	11
Journal	International Statistical Review
Volume	82
Issue number	3
DOIs	https://doi.org/10.1111/insr.12063
Publication status	Published - 1 Dec 2014

Keywords

BUGS
DIC
financial models comparison
mixing representation
normal variance-mean distribution
Variance-Gamma distribution

Access to Document

10.1111/insr.12063

Cite this

@article{5be0a3105c6042d0849933d6553cbc81,

title = "The deviance information criterion in comparison of normal mixing models",

abstract = "Model selection from several non-nested models by using the deviance information criterion within Bayesian inference Using Gibbs Sampling (BUGS) software needs to be treated with caution. This is particularly important if one can specify a model in various mixing representations, as for the normal variance-mean mixing distribution occurring in financial contexts. We propose a procedure to compare goodness of fit of several non-nested models, which uses BUGS software in part.",

keywords = "BUGS, DIC, financial models comparison, mixing representation, normal variance-mean distribution, Variance-Gamma distribution",

author = "Thomas Fung and Wang, \{Joanna J. J.\} and Eugene Seneta",

year = "2014",

month = dec,

day = "1",

doi = "10.1111/insr.12063",

language = "English",

volume = "82",

pages = "411--421",

journal = "International Statistical Review",

issn = "0306-7734",

publisher = "John Wiley \& Sons",

number = "3",

}

TY - JOUR

T1 - The deviance information criterion in comparison of normal mixing models

AU - Fung, Thomas

AU - Wang, Joanna J. J.

AU - Seneta, Eugene

PY - 2014/12/1

Y1 - 2014/12/1

N2 - Model selection from several non-nested models by using the deviance information criterion within Bayesian inference Using Gibbs Sampling (BUGS) software needs to be treated with caution. This is particularly important if one can specify a model in various mixing representations, as for the normal variance-mean mixing distribution occurring in financial contexts. We propose a procedure to compare goodness of fit of several non-nested models, which uses BUGS software in part.

AB - Model selection from several non-nested models by using the deviance information criterion within Bayesian inference Using Gibbs Sampling (BUGS) software needs to be treated with caution. This is particularly important if one can specify a model in various mixing representations, as for the normal variance-mean mixing distribution occurring in financial contexts. We propose a procedure to compare goodness of fit of several non-nested models, which uses BUGS software in part.

KW - BUGS

KW - DIC

KW - financial models comparison

KW - mixing representation

KW - normal variance-mean distribution

KW - Variance-Gamma distribution

UR - https://www.scopus.com/pages/publications/85028095904

U2 - 10.1111/insr.12063

DO - 10.1111/insr.12063

M3 - Article

AN - SCOPUS:85028095904

SN - 0306-7734

VL - 82

SP - 411

EP - 421

JO - International Statistical Review

JF - International Statistical Review

IS - 3

ER -

The deviance information criterion in comparison of normal mixing models

Abstract

Keywords

Access to Document

Other files and links

Fingerprint

Cite this