Bayesian measures of model complexity and fit

By David J. Spiegelhalter, N. G. Best, Bradley P. Carlin, A. van der Linde. 2002

13,118 citations recorded by the source corpus.

Subfields: Artificial Intelligence, Statistics and Probability

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

In Statistics and Probability

Citations in the corpus

A Bayesian model for repeated measures zero-inflated count data with application to outpatient psychiatric service use (2010, methodology): “As an alternative, we consider two computationally tractable Bayesian procedures to compare the fit of competing models: the deviance information criterion (DIC; Spiegelhalter et al., 2002) and a cross-validatory comparison based on an approximation to the marginal likelihood f (y) (Gelfand, 1996).”
Clarifying the effect of toe clipping on frogs with Bayesian statistics (2004, methodology): “Deviance information criteria (DIC; Spiegelhalter et al. 2002) indicating the goodness-of-fit of the original logistic regression of Parris & McCarthy (2001) and the three new models (models A, B and C) Model DIC Logistic regression 532·6 A, constant effect of toe clipping 522·9 B, changing effect…”
Nonlife Ratemaking and Risk Management with Bayesian Additive Models for Location, Scale and Shape (2013, methodology): “As derived by Spiegelhalter et al. (2002), the latter is given by the difference of the posterior mean of the deviance and the deviance of the posterior means, i.e. pD = D(ϑ) − D(ϑ), where D(ϑ) = 1 T T∑ t=1 D ( ϑ(t) ) and ϑ = 1 T T∑ t=1 ϑ(t).”
Markov chain Monte Carlo (MCMC) sampling methods to determine optimal models, model resolution and model choice for Earth Science problems (2009, background): “In addition to the references already cited in this section, we refer the reader to Burnham and Anderson (2002) and Spiegalhalter et al. (2002) for further discussion regarding issues of model choice (the latter is particularly relevant to MCMC sampling methods).”
Use of Maximum Entropy Modeling in Wildlife Research (2009, methodology): “The development of an information-theoretic (e.g., Akaike's Information Criterion [AIC]) approach for Maxent might provide the greatest utility for model selection.”
Advances in degradation modeling : applications to reliability, survival analysis, and finance (2010, neutral): “Martz and Waller [28] provided an early description of the use of Bayesian methods in reliability applications.”
Bayesian quantile regression for ordinal longitudinal data (2016, methodology): “The methods are assessed based on 95% credible intervals and the deviance information criterion (DIC; Spiegelhalter et al., 2002).”
Bayesian Cognitive Modeling: A Practical Course (2014, background): “• DIC gives an option to calculate the Deviance Information Criterion (DIC) statistic (Spiegelhalter et al., 2002).”
Multivariate random parameters zero-inflated negative binomial regression for analyzing urban midblock crashes (2018, neutral): “Deviance information criterion (DIC) is a generalized version of Akaike Information Criterion for evaluating hierarchical models (Spiegelhalter et al., 2002).”
Financial regimes and uncertainty shocks (2019, methodology): “As a reality check on its plausibility, we compare it to a set of simpler alternatives using the Deviance Information Criterion (DIC) of Spiegelhalter et al. (2002).19 We consider five different specifications of the model.”
EXTRACTING MORE OUT OF RELOCATION DATA: BUILDING MOVEMENT MODELS AS MIXTURES OF RANDOM WALKS (2004, background): “13 14 Model Comparison and Goodness of Fit 15 Spiegelhalter et al. (2002) proposed a “Deviance Information Criterion” (DIC) as a 16 natural generalization of Akaike’s Information Criterion (AIC).”
A Comparison of Alternative Models for Testlets (2006, methodology): “Similar to AIC and BIC, the DIC is a model comparison criterion that takes into account both model fit and model complexity, but the DIC provides a model complexity measure that can be applied to hierarchical models (Spiegelhalter et al., 2002).”
A unified approach to model selection using the likelihood ratio test (2011, background): “choice, which is extremely popular in ecology, is to use the Deviance Information Criterion (Spiegelhalter et al. 2002): this is relatively straightforward to calculate (unlike Bayes factors), but it is not with-”