Relevant statistics for Bayesian model choice

  title={Relevant statistics for Bayesian model choice},
  author={J. Marin and N. Pillai and C. Robert and J. Rousseau},
  journal={Quality Engineering},
  • J. Marin, N. Pillai, +1 author J. Rousseau
  • Published 2011
  • Mathematics, Engineering, Biology
  • Quality Engineering
  • The choice of the summary statistics used in Bayesian inference and in particular in ABC algorithms has bearings on the validation of the resulting inference. Those statistics are nonetheless customarily used in ABC algorithms without consistency checks. We derive necessary and sufficient conditions on summary statistics for the corresponding Bayes factor to be convergent, namely to asymptotically select the true model. Those conditions, which amount to the expectations of the summary… CONTINUE READING
    117 Citations

    Figures from this paper

    On Model Selection with Summary Statistics
    • PDF
    Asymptotic properties of approximate Bayesian computation
    • 53
    • Highly Influenced
    • PDF
    On Consistency of Approximate Bayesian Computation
    • 4
    • PDF
    Approximate Bayesian Computation
    • 244
    • PDF
    Bayesian model comparison with un-normalised likelihoods
    • 21
    • PDF
    Bayesian Monte Carlo testing with one-dimensional measures of evidence
    • 1
    • PDF
    Semi-automatic selection of summary statistics for ABC model choice
    • 53
    • Highly Influenced
    • PDF
    On oracle property and asymptotic validity of Bayesian generalized method of moments
    • Cheng Li, Wenxin Jiang
    • Mathematics, Computer Science
    • J. Multivar. Anal.
    • 2016
    • 9
    • PDF
    Robust Approximate Bayesian Inference with Synthetic Likelihood
    • 8
    • PDF


    Constructing summary statistics for approximate Bayesian computation: semi-automatic approximate Bayesian computation (with Discussion)
    • 322
    Semi-automatic Approximate Bayesian Computation
    • 29
    • Highly Influential
    Approximate Bayesian computation scheme for parameter inference and model selection in dynamical systems
    • 1,113
    • PDF
    ABC likelihood-free methods for model choice in Gibbs random fields
    • 148
    • PDF
    Likelihood-free estimation of model evidence
    • 109
    • PDF
    Lack of confidence in approximate Bayesian computation model choice
    • 278
    • PDF
    Bayesian estimation of quantile distributions
    • 54
    Bayesian measures of model complexity and fit
    • 10,036
    • PDF
    Asymptotic behaviour of the posterior distribution in overfitted mixture models
    • 187
    • PDF
    Approximating Interval hypothesis : p-values and Bayes factors
    • 32