Marginal Likelihood and Bayes Factors for Dirichlet Process Mixture Models

By Sanjib Basu, Siddhartha Chib. 2003

151 citations recorded by the source corpus.

Subfields: Artificial Intelligence, Statistics and Probability

View Paper: https://www.semanticscholar.org/paper/9d42849ccc3f94cdcaa376ca6004c85ec0b284d3

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

Functional clustering by Bayesian wavelet methods (2006, methodology): “Ideally, the marginal likelihood can be calculated for the DP priors by using the collapsed sequential importance sampling methods of Basu and Chib (2003).”
Nonparametric Bayesian Modeling for Multivariate Ordinal Data (2005, methodology): “Regarding formal model comparison, Basu and Chib (2003) discussed the use of Bayes factors for DP mixture models.”
Inference for Mixtures of Finite Polya Tree Models (2006, methodology): “Data on n = 53 prostate cancer patients were used for illustrative purposes by Chib (1995) and Basu and Chib (2003). These data were fit to a Bernoulli regression model with a random PT link centered at the probit and logistic curves.”
Bayesian Nonparametric Spatial Modeling With Dirichlet Process Mixing (2005, background): “One possibility is to attempt to calculate Bayes factors, following the approach laid out by Basu and Chib (2003). Of course, one might prefer not to reduce a model to a single number to make comparisons.”
Two-sample Bayesian Nonparametric Hypothesis Testing (2009, methodology): “The calculation of Bayes factors for Dirichlet process-based models is discussed by Basu and Chib (2003). Goodness of fit testing using mixtures of triangular distributions is considered by McVinish et al. (2009). An alternative form of prior, the Pólya tree, was considered by Berger and Guglielmi (”
Accept–reject Metropolis–Hastings sampling and marginal likelihood estimation (2005, methodology): “The proposed method joins a substantial literature concerned with the estimation of marginal likelihoods and Bayes factors (e.g. Newton and Raftery, 1994; Gelfand and Dey, 1994; Carlin and Chib, 1995; Chib, 1995; Green, 1995; Verdinelli and Wasserman, 1995; Meng and Wong, 1996; Chen and Shao, 1997; ”
Estimating the Integrated Likelihood via Posterior Simulation Using the Harmonic Mean Identity (2006, methodology): “Bickel and Ghosh (1990) proposed such an expansion where the leading term is of the form (10). This expansion would not in itself be useful for the present purpose because the leading term still yields an infinite log integrated likelihood, but the basic idea may be fruitful in a modified form. Othe”
Fast Bayesian Inference in Dirichlet Process Mixture Models (2011, background): “Typical MCMC algorithms do not allow such comparisons, as marginal likelihoods are not estimated, though there has been some recent work to address this gap (Basu and Chib 2003).”
Modelling of extreme wave heights and periods through copulas (2005, background): “There exists a large literature on the Dirichlet Process (DP) and the DP prior became popular for estimating the random effects in mixed linear models (see Basu and Chib, 2001).”
Semiparametric Bayes analysis of longitudinal data treatment models (2002, methodology): “It is clear that this posterior density does not belong to a known family of distributions but the sampling strategy that is developed by Chib and Greenberg (1998) in a related context can be used.”