METHODS OF VARIABLE SELECTION IN REGRESSION MODELING

How researchers see this paper

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

AutoScore: A Machine Learning–Based Automatic Clinical Score Generator and Its Application to Mortality Prediction Using Electronic Health Records (2020, methodology): “Third, we use a parsimony plot (model performance vs complexity) to determine the appropriate number of variables ( m ), balancing the trade-off between performance and sparsity [39,40].”
MMI: Multimodel inference or models with management implications? (2015, background): “…hypothesis tests or making multiple comparisons from an analysis of variance model; the probability of making a Type I error increases with the number of tests performed or, in the present case, with the number of potential explanatory variables considered (Murtaugh 1998, Whittingham et al. 2006).”
Bootstrap Methods for Developing Predictive Models (2004, background): “Murtaugh reported that the probability of correctly identifying variables was inversely proportional to the number of variables under consideration (Murtaugh 1998).”
Automated variable selection methods for logistic regression produced unstable models for predicting acute myocardial infarction mortality. (2004, methodology): “Omitting important prognostic factors results in a systematic mis-estimation of the regression coefficients and biased prediction, and including too many predictors results in the loss of precision in the estimation of the regression coefficients and the predictions of new responses [2].”