Bayesian and Conditional Frequentist Testing of a Parametric Model Versus Nonparametric Alternatives

Berger, James O.; Guglielmi, Alessandra

doi:10.1198/016214501750333045

Cited by 122 publications

(90 citation statements)

References 11 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Mixtures of Polya trees were developed by Berger and Guglielmi (2001), Hanson and Johnson (2002), and Hanson (2006). See also Paddock et al (2003) for related work.…”

Section: Prior For Baseline Survivalmentioning

confidence: 99%

Predictive comparison of joint longitudinal-survival modeling: a case study illustrating competing approaches

2010

View full text Add to dashboard Cite

The joint modeling of longitudinal and survival data has received extraordinary attention in the statistics literature recently, with models and methods becoming increasingly more complex. Most of these approaches pair a proportional hazards survival with longitudinal trajectory modeling through parametric or nonparametric specifications. In this paper we closely examine one data set previously analyzed using a two parameter parametric model for Mediterranean fruit fly (medfly) egg-laying trajectories paired with accelerated failure time and proportional hazards survival models. We consider parametric and nonparametric versions of these two models, as well as a proportional odds rate model paired with a wide variety of longitudinal trajectory assumptions reflecting the types of analyses seen in the literature. In addition to developing novel nonparametric Bayesian methods for joint models, we emphasize the importance of model selection from among joint and non joint models. The default in the literature is to omit at the outset non joint models from consideration. For the medfly data, a predictive diagnostic criterion suggests that both the choice of survival model and longitudinal assumptions can grossly affect model adequacy and prediction. Specifically for these data, the simple joint model used in by Tseng et al. (Biometrika 92:587-603, 2005) and models with much more flexibility in their longitudinal components are predictively outperformed by simpler analyses. This case study underscores the need for data analysts to compare on the basis of predictive performance different joint models and to include non joint models in the pool of candidates under consideration.

show abstract

“…Mixtures of Polya trees were developed by Berger and Guglielmi (2001), Hanson and Johnson (2002), and Hanson (2006). See also Paddock et al (2003) for related work.…”

Section: Prior For Baseline Survivalmentioning

confidence: 99%

Predictive comparison of joint longitudinal-survival modeling: a case study illustrating competing approaches

2010

View full text Add to dashboard Cite

show abstract

“…Various priors have been used for goodness of fit problems in the literature including Dirichlet processes (Carota & Paramigiani, 1996), Polya trees (Berger & Guglielmi, 2001) and infinite dimensional exponential families (Verdinelli & Wasserman, 1998). With priors such as these, verifying that the general conditions given in section 2 hold is far from trivial.…”

Section: Introductionmentioning

confidence: 99%

Bayesian Goodness of Fit Testing with Mixtures of Triangular Distributions

McVinish

Rousseau

Mengersen

2009

Scandinavian J Statistics

View full text Add to dashboard Cite

show abstract

“…Finally, for appropriate model comparisons, it is desirable, if possible, to match the prior speci cations in the two models, at least for similar parameters (see Berger and Guglielmi 2001 for further discussion). If we are comparing a DPM model with another nonparametric model, then this issue needs to be taken up on a case-by-case basis, as shown in our examples.…”

Section: Model Comparison Problemmentioning

confidence: 99%

“…It is important to bear in mind that, despite the similarity in nomenclature, the MDP speci cation is rather different from the DPM model that we consider in the sequel. In more recent work, Berger and Guglielmi (2001) modeled the nonparametric alternative by a Pólya tree process and computed the Bayes factor for a default reference prior, and Ishwaran, James, and Sun (2001) compared models with different number of unique mixture components by subsuming the models within a nite mixture model.…”

Section: Introductionmentioning

confidence: 99%

Marginal Likelihood and Bayes Factors for Dirichlet Process Mixture Models

Basu

Chib

2003

Journal of the American Statistical Association

123

127

View full text Add to dashboard Cite

We present a method for comparing semiparametric Bayesian models, constructed under the Dirichlet process mixture (DPM) framework, with alternative semiparameteric or parameteric Bayesian models. A distinctive feature of the method is that it can be applied to semiparametric models containing covariates and hierarchical prior structures, and is apparently the rst method of its kind. Formally, the method is based on the marginal likelihood estimation approach of Chib (1995) and requires estimation of the likelihood and posterior ordinates of the DPM model at a single high-density point. An interesting computation is involved in the estimation of the likelihood ordinate, which is devised via collapsed sequential importance sampling. Extensive experiments with synthetic and real data involving semiparametric binary data regression models and hierarchical longitudinal mixed-effects models are used to illustrate the implementation, performance, and applicability of the method.

show abstract

Bayesian and Conditional Frequentist Testing of a Parametric Model Versus Nonparametric Alternatives

Cited by 122 publications

References 11 publications

Predictive comparison of joint longitudinal-survival modeling: a case study illustrating competing approaches

Predictive comparison of joint longitudinal-survival modeling: a case study illustrating competing approaches

Bayesian Goodness of Fit Testing with Mixtures of Triangular Distributions

Marginal Likelihood and Bayes Factors for Dirichlet Process Mixture Models

Contact Info

Product

Resources

About