The Variable Selection Problem

By Edward I. George. 2000

386 citations recorded by the source corpus.

View Paper: https://www.semanticscholar.org/paper/7424ba38c001f656fa0946fc05977dc6fe3cec47

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

Introduction to the LASSO (2018, background): “There is a large literature on variable selection methods for linear models (see [9] and [10]) and for high dimensional problems (see [11] and [12]).”
Feature subset selection for logistic regression via mixed integer optimization (2016, background): “where F is a preset dimensionality penalty [13].”
Detection of Multiple Change Points from Clustering Individual Observations (2002, neutral): “See George (2001) for further discussion.”
Data Mining for Molecules with 2-D Neural Network Sensitivity Analysis (2003, background): “Edward George presents a good summary of the history and advances in variable selection for linear problems (George 1999).”
Polynomial nonlinear spatio‐temporal integro‐difference equation models (2011, methodology): “…the state-dependent discrete-space representation of the nonlinear AR process eqn (1) is given by the nonlinear stochastic IDE, YtðsÞ ¼ Z D ksðu; hkÞgðYt 1ðuÞ; hgÞduþ gtðsÞ; s;u 2 D; t 2 T ; ð3Þ where g(Æ; hg) is a nonlinear function in fYt 1(u) : u 2 Dg and hg, and ks(u; hk) is the kernel that…”
Efficient Empirical Bayes Variable Selection and Estimation in Linear Models (2005, methodology): “George and Foster (2000) showed that these criteria correspond to a hierarchical Bayes model selection procedure under a particular class of priors. This gives a new perspective on various earlier model selection methods and puts them in a unified framework. The hierarchical Bayes formulation puts a”
Compatibility of prior specifications across linear models (2008, background): “viewpoint; see, for example, George (1999) and Robert (2001, Chapter 7). In particular, the problem of variable selection in linear models has received considerable attention; see the review paper of George (2000) and Guido Consonni is Professor, Dipartimento di Economia Politica e Metodi Quantitati”
Estimating brain functional connectivity with sparse multivariate autoregression (2005, methodology): “A very popular method is stochastic search variable selection (SSVS) (George & McCulloch 1997; George 2000).”
Objective Testing Procedures in Linear Models: Calibration of the p‐values (2006, methodology): “The hyperparameter g is sometimes chosen as g =n (Shively et al., 1999), sometimes g is estimated by empirical Bayes methods (Clyde & George, 2000; George & Foster, 2000), and a hyperprior for g has also been considered (Clyde, 2001).”
BAYESIAN LINEAR REGRESSION WITH SPARSE PRIORS (2014, background): “2) were considered by many authors, including [8, 20, 21, 26, 34, 41, 48].”
Agent-based Models of Geographical Systems (2012, background): “, & Georgeff, M. P. (1995). BDI-agents: From theory to practice . In Proceedings of the First International Conference on Multi-Agents-Systems . Menlo: AAAI Press. Scherer, K. R. (1999). Appraisal theory. In T. Dalgleish & M. Power (Eds.), Handbook of cognition and emotion (pp. 637–663). New York: W”
Combining Linear Regression Models (2005, background): “For such cases, the role and meaning of the calculated posterior probabilities of the models are not quite clear, and examining frequentist properties of the BMA methods is viewed as valuable by many (cf. George 2000).”
The Effective Sample Size (2014, methodology): “Other model selection priors which make similar use of n can be found in Bayarri and García-Donato (2007, 2008), Bayarri et al. (2009), George (2000), Liang et al. (2008), Verdinelli and Wasserman (1995), and Zellner (1986).”
Deep learning for short-term traffic flow prediction (2016, methodology): “Therefore, the objective function dropout is equivalent to a Bayes ridge regression with a g -prior George [2000]. The end model has minimal deviance for a validation data set.”