Bayesian Model Selection and Model Averaging.

By Larry A. Wasserman. Journal of Mathematical Psychology, 2000

1,002 citations recorded by the source corpus.

View Paper: https://www.semanticscholar.org/paper/22d81c097f97b237ea5613064017faebf66c3398

How researchers see this paper

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

In Statistics and Probability

Citations in the corpus

AIC model selection using Akaike weights (2004, methodology): “A popular alternative model selection criterion is the Bayesian information criterion or BIC (e.g., Burnham & Anderson, 2002; Hastie, Tibshirani, & Friedman, 2001; Kass & Raftery, 1995; Schwarz, 1978; Wasserman, 2000).”
Model averaging in ecology: a review of Bayesian, information-theoretic, and tactical approaches for predictive inference (2018, methodology): “eory Bayes’ formula can be applied to choosing among models in much the same 396 way as to parameter values (Wasserman, 2000).”
The Practical Implementation of Bayesian Model Selection (2001, background): “12) may be considered an implicit approximation to p(Y \ M) under a "unit information prior" (Kass and Wasserman 1995) or under a "normalized" Jeffreys prior (Wasserman 2000).”
Growth Empirics and Reality (2001, background): “As is well known (see Wasserman 2000, p. 94 for a nice exposition), when Bayesian model selection between two models is based on posterior odds ratios and the prior odds on the models are equal, the posterior odds ratios will equal the ratio of their likelihoods, that is, the posterior odds will…”
On comparing three artificial neural networks for wind speed forecasting (2010, methodology): “In this case, the Bayesian model averaging (BMA) method [33], an adaptive model combination method, might be used for post-processing different plausible forecasts to form a single forecast.”
Comparison of Akaike information criterion (AIC) and Bayesian information criterion (BIC) in selection of an asymmetric price relationship (2010, methodology): “Essentially, it involves the use of a model selection criterion to find the best fitting model to the data (Wasserman, 2000).”
Inverse groundwater modeling for hydraulic conductivity estimation using Bayesian model averaging and variance window (2008, methodology): “…and Bayesian model averaging (BMA) [Draper, 1995; Madigan and Raftery, 1994; Raftery, 1995; Hoeting et al., 1999; Brown et al., 2002] are naturally developed and linked through the posterior model probability Pr(q(p)jD) that represents the weight of a model against other models [Wasserman, 2000].”
A Survey on ensemble learning under the era of deep learning (2021, methodology): “For the wind power forecasting, Wang et al. (Wang et al. 2020) proposed a hybrid approach named BMA-EL based on Bayesian model averaging (BMA) (Hoeting et al. 1999; Wasserman 2000) and ensemble learning (EL).”
Exploring patterns of Latino/a children's school readiness at kindergarten entry and their relations with Grade 2 achievement (2013, positive): “The value of the ratio is then compared to the Jeffery’s Scale of Evident (Wasserman, 2000), for which 1 < BF < 3 is considered weak evidence for Model K over Model K +1, 3 < BF < 10 is moderate evidence for Model K, and BF > 10 is strong evidence for Model K.”
Schwarz, Wallace, and Rissanen: Intertwining Themes in Theories of Model Selection (2001, background): “In some cases, it is not necessary to pick a specific model order; it may be better, in fact, to employ all the models in a weighted fashion (Wasserman, 2000; Hoeting et al., 1999; Chickering & Heckerman, 2000), as Singer & Feder (1999) illustrate for linear least-squares prediction problems.”
Model selection and model averaging in phylogenetics: advantages of akaike information criterion and bayesian approaches over likelihood ratio tests. (2004, background): “Although in general model selection is concerned with the selection of just the best fit model, Bayesian approaches allow us to make inferences based on the entire set of candidate models, or model averaging (Hoeting et al., 1999; Madigan and Raftery, 1994; Raftery, 1996; Wasserman, 2000).”
Bayesian and LASSO Regressions for Comparative Permeability Modeling of Sandstone Reservoirs (2018, background): “The optimal model has the maximum posterior probability, and minimum BIC.”
Diffusion Posterior Sampling for General Noisy Inverse Problems (2022, methodology): “Given the succinct representation of (10), our proposal is to use the Laplace approximation (Wasserman, 2000) to compute the intractable integration in (7).”
Multimodel Inference (2004, result): “…that has evaluated, or even considered, model averaged inferences compared to the best model strategy that model averaging is superior (e.g., Buckland et al. 1997; Hoeting et al. 1999; Wasserman 2000; Breiman 2001; Burnham and Anderson 2002; Hansen and Kooperberg 2002; Lukacs et al. in review).”
A Bayesian information criterion for singular models (2013, methodology): “3 in Watanabe (2009). Note that for j < i this bound is strictly smaller than dim(Mi)/2 = (3i− 1)/2.”
Do People Make Decisions Under Risk Based on Ignorance? An Empirical Test of the Priority Heuristic Against Cumulative Prospect Theory (2008, methodology): “Figure 2 indicates that the specific prediction of the PH – namely that the certainty effect would disappear if the minimum gains did not differ considerably – was not supported by the data.”