The Relationship Between Variable Selection and Data Agumentation and a Method for Prediction

By D. M. Allen. 1974

How researchers see this paper

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

In Optimization & Training

Citations in the corpus

True Few-Shot Learning with Language Models (2021, methodology): “Cross-Validation (CV) [22–24] is one of the most widely used methods for estimating generalization loss [25].”
CONSISTENCY OF CROSS VALIDATION FOR COMPARING REGRESSION PROCEDURES (2007, background): “1. Introduction. Cross validation (e.g., Allen [ 2 ], Stone [25] and Geisser [9]) is one of the most commonly used model selection criteria.”
Calibrating the Degrees of Freedom for Automatic Data Smoothing and Effective Curve Checking (2003, methodology): “The most frequently used procedure for bandwidth selection is cross-validation (CV) (Allen 1974; Stone 1974).”
An Introduction to Kernel and Nearest-Neighbor Nonparametric Regression (1992, methodology): “A popular heuristic is leave-one-out cross-validation, or PRESS (prediction sum of squares) (Allen 1974; Geisser 1975; Stone 1974) Case i is deleted from the data for predicting p(t,) giving the estimate P-,(ti) = C w j ~ j / C "'j, j t i j t i where the weights are defined as in Section The…”
The Design and Analysis of Computer Experiments (2003, background): “Important early references describing cross-validation are Allen (1974), Stone (1974), and Stone (1977); Hastie et al (2001) summarize recent applications.”
A survey of cross-validation procedures for model selection (2009, unknown): “Leave-one-out (LOO, Stone, 1974; Allen, 1974; Geisser, 1975) is the most classical exhaustive CV procedure, corresponding to the choice n t = n 1 : Each data point is successively “left out” from the sample and used for validation.”
On Optimal Generalizability in Parametric Learning (2017, methodology): “Hence, provided that the regularized loss function is quadratic in θ , ε The fact that the cross validation vector could be obtained for kernel ridge regression in closed form without actually performing cross validation is not new, and the method is known as PRESS [11].”
Robust Locally Weighted Regression and Smoothing Scatterplots (1979, methodology): “In such cases the PRESS procedure of Allen (1974), used ordinarily for choosing a subset of the independent variables in a regression, can be tailored to robust locally weighted regression to choose f .”
Estimation Stability With Cross-Validation (ESCV) (2013, methodology): “More commonly used today are model-free approaches like cross-validation ( CV ) (Allen 1974; Stone 1974) and bootstrap methods (Efron 1979; Zhang 1993; Shao 1996).”
Jackknife model averaging (2012, methodology): “(Delete-one) cross-validation for selection of regression models was introduced by Allen (1974), Stone (1974), Geisser (1974) and Wahba and Wold (1975), and its optimality demonstrated by Li (1987) for homoskedastic regression and Andrews (1991) for heteroskedastic regression.”
Logistic regression and artificial neural network classification models: a methodology review (2002, unknown): “This process of n-fold cross-validation builds n models; the numbers reported are the averages over all n test sets [18,19].”
Model Selection Techniques: An Overview (2018, methodology): “Cross-validation (CV) [38], [39] is a class of model selection methods widely used in machine learning practice.”
Cross-Validation With Confidence (2017, methodology): “Cross-validation [27, 1, 15] is one of the most popular methods for model and tuning parameter selection in statistics and machine learning, due to its conceptual simplicity and wide applicability.”
Information-theoretic approaches to statistical analysis in behavioural ecology: an introduction (2010, methodology): “These model selection methods include, for example, the Mallows’ Cp method (Mallows 1973), the Bayes information criterion (Schwarz 1978), the final prediction error method (Shibata 1981), the generalised information criterion (Rao and Wu 1989) and its analogues (Potscher 1989), the delete-one cross”
Torturing data for the sake of generality: How valid are our regression models? (2000, methodology): “Cross validation was proposed originally for this reason (Allen, 1974; Stone, 1974; Geisser, 1975) and involves the splitting of the data into two groups of observations, with one group being used in model construction (often called the training sample) and the other group reserved for model…”
Model selection for ecologists: the worldviews of AIC and BIC. (2014, methodology): “…consistent estimators, namely BIC, Hannan and Quinn information (Hannan and Quinn 1979), and GM (Geweke and Meese 1981), and efficient estimators, namely AIC, Mallows’ Cp (Mallows 1973), predicted residual sum of squares (PRESS; Allen 1974), Akaike’s FPE (Akaike 1969), and cross validation.”