Bayesian model averaging

By Tomohiro Ando. 2010

272 citations recorded by the source corpus.

Subfields: Control and Systems Engineering, Ocean Engineering, Statistics and Probability

View Paper: https://www.semanticscholar.org/paper/05a3e083b69efec0885a5363aee813643d68f22e

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Control and Systems Engineering

In Ocean Engineering

In Statistics and Probability

Citations in the corpus

Using Stacking to Average Bayesian Predictive Distributions (with Discussion) (2017, methodology): “This is called Bayesian model averaging (BMA) and is optimal if the prior is correct, that is, if the method is evaluated based on its frequency properties evaluated over the joint prior distribution of the models and their internal parameters (Madigan et al., 1996; Hoeting et al., 1999).”
Decoupling Shrinkage and Selection in Bayesian Linear Models: A Posterior Summary Perspective (2014, methodology): “(4) An early reference adopting this formulation is Raftery, Madigan, and Hoeting (1997); see also Clyde and George (2004).”
Practical Bayesian model evaluation using leave-one-out cross-validation and WAIC (2015, methodology): “After ﬁtting a Bayesian model we often want to measure its predictive accuracy, for its own sake or for purposes of model comparison, selection, or averaging (Geisser and Eddy, 1979, Hoeting et al., 1999, Vehtari and Lampinen, 2002, Ando and Tsay, 2010, Vehtari and Ojanen, 2012).”
Robust Feature Selection Using Ensemble Feature Selection Techniques (2008, methodology): “Variation in the feature selectors can be achieved by various methods: choosing different feature selection techniques, instance level perturbation, feature level perturbation, stochasticity in the feature selector, Bayesian model averaging, or combinations of these techniques [8,9].”
Bayesian mixed-effects inference on classification performance in hierarchical data sets (2012, methodology): “The second option to make use of the model evidences of competing models is Bayesian model averaging (Cooper and Herskovits, 1992; Madigan and Raftery, 1994; Madigan et al., 1996).”
Model Averaging and Its Use in Economics (2017, methodology): “For dynamic models, an efficient implementation of the DMA methodology of Raftery et al. (2010) is provided in the R package eDMA (Catania and Nonejad, 2017b), as described in Catania and Nonejad (2017a).”
Model averaging in ecology: a review of Bayesian, information-theoretic, and tactical approaches for predictive inference (2018, methodology): “A problem for the computation of model weights when performing Bayesian inference across multiple models is the influence of the choice of parameter priors, especially “uninformative” ones (see Chickering and Heckerman 1997, Hoeting et al. 1999: section 5).”
Inconsistency of Bayesian Inference for Misspecified Linear Models, and a Proposal for Repairing It (2014, methodology): “Adopting a different prior (such as the g-prior) does not help, with two exceptions in model averaging: (a) if using a prior that follows the suggestions of Raftery et al. (1997), then Bayes works quite well with 50% ‘easy’ points, but starts to fail dramatically again (in contrast to SafeBayes) onc”
Optimal Weight Choice for Frequentist Model Average Estimators (2011, background): “Reviews of the relevant Bayesian literature can be found in the works of Hoeting et al. (1999), Raftery, Madigan, and Hoeting (1997).”
Socio-economic development and electricity access in developing economies: A long-run model averaging approach (2019, background): “Javadi et al. (2013) found that rural electrification has a positive impact on education and health.”
Bayesian model selection for group studies - Revisited (2014, unknown): “(7) is a direct application of Bayesian model averaging (Madigan et al., 1996), where we have averaged over H0 and H1.”
DEM: A variational treatment of dynamic systems (2008, methodology): “The objective is to compute q(ϑ) for each model by maximising the free-energy and then use F ≈ ln p( y|m) as a lower-bound approximation to the log-evidence for model comparison (e.g., Penny et al., 2004) or averaging (e.g., Trujillo-Barreto et al., 2004).”
Scalable Variational Inference for Bayesian Variable Selection in Regression, and Its Accuracy in Genetic Association Studies (2012, methodology): “Among the variety of approaches to variable selection for regression, the Bayesian approach (George and McCulloch 1997; Raftery, Madigan, and Hoeting 1997) stands out because we can assess the predictive value of a variable X i simply by computing the posterior probability that it is included in…”
Stacking for Non-mixing Bayesian Computations: The Curse and Blessing of Multimodal Posteriors (2020, background): “Proofs and related lemmas are in Appendix A.”