Methods and Criteria for Model Selection

By Joseph B. Kadane, Nicole A. Lazar. 2004

442 citations recorded by the source corpus.

Model selection is an important part of any statistical analysis and, indeed, is central to the pursuit of science in general. Many authors have examined the question of model selection from both frequentist and Bayesian perspectives, and many tools for selecting the “best model” have been suggested in the literature. This paper considers the various proposals from a Bayesian decision–theoretic perspective.

Subfields: Statistics and Probability

View Paper: https://www.semanticscholar.org/paper/5b8553645f20e1ce24accbf136146e0e098e44ec

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

Model selection and psychological theory: A discussion of the differences between the Akaike Information Criterion (AIC) and the Bayesian Information Criterion (BIC) (2012, background): “The parametricness index is an example of adaptive model selection criteria, a more recent movement in statistical theory where the choice of a model selection criterion is itself data-driven (Kadane & Lazar, 2004).”
Energy dissipation and damage evolution analyses for the dynamic compression failure process of red-sandstone after freeze-thaw cycles (2017, background): “R(2) (Kadane and Lazar, 2004) is amodification of R(2) that adjusts for the number of explanatory terms in a model when taking the sample size into consideration.”
Stationary and nonstationary generalized extreme value modelling of extreme precipitation over a mountainous area under climate change (2014, methodology): “This agrees with the well-known tendency of the BIC to select more parsimonious models than the AIC (e.g. Kadane and Lazar, 2004).”
Inference for Mixtures of Finite Polya Tree Models (2006, methodology): “Kadane and Lazar (2004) noted that predictive measures such as the pseudo Bayes factor attempt to find “which models give the best predictions of future observations generated from the same process as the original data,” whereas Bayes factors attempt to find “which models best explain the observed…”
Model Selection Techniques: An Overview (2018, background): “There have been many overview papers on model selection scattered in the communities of signal processing [1], statistics [2], machine learning [3], epidemiology [4], chemometrics [5], ecology and evolution [6].”
A review of automatic selection methods for machine learning algorithms and hyper-parameter values (2016, background): “2013)] or nested models (for frequentist statistics) (Claeskens and Hjort 2008; Kadane and Lazar 2004).”
Model recommendation: Generating object detectors from few samples (2015, methodology): “This is inspired by the M -closed framework [21] for statistical evaluation.”
Privacy Expectations and Preferences in an IoT World (2017, methodology): “We eliminated the variables with the largest p-value in each step of the model selection and continued the elimination until the BIC reached the global minimum [15].”