Compatibility of Prior Specifications Across Linear Models
Consonni, Guido ; Veronese, Piero
Statist. Sci., Tome 23 (2008) no. 1, p. 332-353 / Harvested from Project Euclid
Bayesian model comparison requires the specification of a prior distribution on the parameter space of each candidate model. In this connection two concerns arise: on the one hand the elicitation task rapidly becomes prohibitive as the number of models increases; on the other hand numerous prior specifications can only exacerbate the well-known sensitivity to prior assignments, thus producing less dependable conclusions. Within the subjective framework, both difficulties can be counteracted by linking priors across models in order to achieve simplification and compatibility; we discuss links with related objective approaches. Given an encompassing, or full, model together with a prior on its parameter space, we review and summarize a few procedures for deriving priors under a submodel, namely marginalization, conditioning, and Kullback–Leibler projection. These techniques are illustrated and discussed with reference to variable selection in linear models adopting a conventional g-prior; comparisons with existing standard approaches are provided. Finally, the relative merits of each procedure are evaluated through simulated and real data sets.
Publié le : 2008-08-15
Classification:  Bayes factor,  compatible prior,  conjugate prior,  g-prior,  hypothesis testing,  Kullback–Leibler projection,  nested model,  variable selection
@article{1233153062,
     author = {Consonni, Guido and Veronese, Piero},
     title = {Compatibility of Prior Specifications Across Linear Models},
     journal = {Statist. Sci.},
     volume = {23},
     number = {1},
     year = {2008},
     pages = { 332-353},
     language = {en},
     url = {http://dml.mathdoc.fr/item/1233153062}
}
Consonni, Guido; Veronese, Piero. Compatibility of Prior Specifications Across Linear Models. Statist. Sci., Tome 23 (2008) no. 1, pp.  332-353. http://gdmltest.u-ga.fr/item/1233153062/