Department of Mathematics
 Search | Help | Login | pdf version | printable version

Math @ Duke



Publications [#257956] of David B. Dunson


Papers Published

  1. Mitra, R; Dunson, D, Two-level stochastic search variable selection in GLMs with missing predictors., The International Journal of Biostatistics, vol. 6 no. 1 (January, 2010), pp. Article-33, ISSN 1557-4679 [doi]
    (last updated on 2019/05/24)

    Stochastic search variable selection (SSVS) algorithms provide an appealing and widely used approach for searching for good subsets of predictors while simultaneously estimating posterior model probabilities and model-averaged predictive distributions. This article proposes a two-level generalization of SSVS to account for missing predictors while accommodating uncertainty in the relationships between these predictors. Bayesian approaches for allowing predictors that are missing at random require a model on the joint distribution of the predictors. We show that predictive performance can be improved by allowing uncertainty in the specification of predictor relationships in this model. The methods are illustrated through simulation studies and analysis of an epidemiologic data set.
ph: 919.660.2800
fax: 919.660.2821

Mathematics Department
Duke University, Box 90320
Durham, NC 27708-0320