2014
DOI: 10.1111/rssb.12062
|View full text |Cite
|
Sign up to set email alerts
|

The Deviance Information Criterion: 12 Years on

Abstract: Summary The essentials of our paper of 2002 are briefly summarized and compared with other criteria for model comparison. After some comments on the paper's reception and influence, we consider criticisms and proposals forimprovement made by us and others.

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
1
1
1
1

Citation Types

3
456
0
3

Year Published

2014
2014
2022
2022

Publication Types

Select...
9
1

Relationship

0
10

Authors

Journals

citations
Cited by 562 publications
(462 citation statements)
references
References 26 publications
3
456
0
3
Order By: Relevance
“…whether the extended model is favored as compared to ΛCDM. To aid in this aim, we follow Joudaki et al (2017) in using the Deviance Information Criterion (DIC; Spiegelhalter, Best & Carlin 2002, also see Kunz, Trotta & Parkinson 2006, Liddle 2007, Trotta 2008, and Spiegelhalter et al 2014, given by the sum of two terms:…”
Section: Model Selection and Dataset Concordancementioning
confidence: 99%
“…whether the extended model is favored as compared to ΛCDM. To aid in this aim, we follow Joudaki et al (2017) in using the Deviance Information Criterion (DIC; Spiegelhalter, Best & Carlin 2002, also see Kunz, Trotta & Parkinson 2006, Liddle 2007, Trotta 2008, and Spiegelhalter et al 2014, given by the sum of two terms:…”
Section: Model Selection and Dataset Concordancementioning
confidence: 99%
“…For all indices and seasons, the DIC favoured the climate-informed models over the classical distribution for a larger number of stations compared to the slope 375 significance. DIC has received some criticism for not adequately penalising complex models and tending to choose overfitted models (Silva et al, 2017;Spiegelhalter et al, 2014). Our results show that at least compared to the slope significance, DIC is a weaker criterion for model selection.…”
mentioning
confidence: 81%
“…Initial parameter starting values are estimated using maximum-likelihood methods, these starting values are then used in Bayesian Markov Chain Monte Carlo estimation, run for 50,000 iterations, confirmed as adequate according to Raftery-Lewis diagnostics (Browne, 2016). The Deviance Information Criterion (DIC) is used to compare the fit of models; similar to likelihood-based criterions like the AIC, models with smaller DIC values are preferred (Spiegelhalter et al, 2014).…”
Section: Equationmentioning
confidence: 99%