Prediction error estimation: a comparison of resampling methods

By Annette M. Molinaro, Richard M. Simon, Ruth M. Pfeiffer. Bioinform., 2005

1,299 citations recorded by the source corpus.

View Paper: https://www.semanticscholar.org/paper/7dbb229a740d48f7791ef21e3f6b51d6090b905e

How researchers see this paper

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

In Genetics

In Molecular Biology

Citations in the corpus

An adaptive algorithm to build up sparse polynomial chaos expansions for stochastic finite element analysis (2010, background): “A comparative study reported in [32] indicates that this technique generally performs very well in terms of estimation bias and mean-square error.”
Evaluating Random Forests for Survival Analysis using Prediction Error Curves (2012, background): “The .632+ was studied for other loss functions including the Brier score (Molinaro, Simon, and Pfeiﬀer 2005; Jiang and Simon 2007; Wehberg and Schumacher 2004; Gerds and Schumacher 2007).”
Selecting Representative Data Sets (2012, neutral): “Data splitting methods and algorithms and their comparison can be found in literature [15, 68, 83, 86].”
Estimating classification error rate: Repeated cross-validation, repeated hold-out and bootstrap (2009, methodology): “The estimators obtained by two methods were compared in the vast literature (Efron, 1983; Crawford, 1989; Fitzmaurice et al., 1991; Efron and Tibshirani, 1997; Braga-Neto and Dougherty, 2004; Wehberg and Schumacher, 2004; Molinaro et al., 2005; Schiavo and Hand, 2000).”
Applied Predictive Modeling (2013, background): “Research (Molinaro 2005; Kim 2009) indicates that repeating k-fold crossvalidation can be used to effectively increase the precision of the estimates while still maintaining a small bias.”
A survey of cross-validation procedures for model selection (2009, neutral): “On realistic synthetic and real biological data, Molinaro et al. (2005) compared the bias of LOO, VFCV and .632+ bootstrap: The bias decreases with n t , and is generally minimal for LOO.”
Towards a biogeographic regionalization of the European biota (2010, methodology): “We used k-means clustering (Han & Kamber, 2001; Bishop, 2005) combined with v-fold cross-validation (e.g. Molinaro et al., 2005) to generate biotic regions for each taxon.”
Hyperparameter optimization: Foundations, algorithms, best practices, and open challenges (2021, background): “A typical instantiation that is often recommended is 10 times repeated 10-fold CV (Molinaro et al., 2005).”
Artificial Intelligence in Cardiovascular Imaging (2019, methodology): “This would be referred to as 10-fold cross-validation and has been shown to have smaller bias for discriminant analysis than a traditional splitsample approach (training and testing) (24).”
Bias in error estimation when using cross-validation for model selection (2006, background): “However, low variance methods can also have a large inherent bias [12] (either positive or negative).”
Adaptive sparse polynomial chaos expansion based on least angle regression (2011, background): “A comparative study reported in [45] indicates that the leave-one-out technique generally performs very well in terms of estimation bias and mean-square error.”
Deep Learning for Prediction of Obstructive Disease from Fast Myocardial Perfusion SPECT: A Multicenter Study (2018, neutral): “…a stratified 10-fold cross-validation procedure to reduce the variance in the prediction error, maximize the use of data while preventing model overfitting, and guard against testing hypotheses suggested by arbitrarily split data, a known limitation of the single split validation approach (24–26).”