Prognostic Modeling with Logistic Regression Analysis

By Ewout W. Steyerberg, Marinus J. C. Eijkemans, Frank E. Harrell, J. Dik F. Habbema. 2001

656 citations recorded by the source corpus.

Subfields: Control and Systems Engineering, Statistics and Probability

How researchers see this paper

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

In Control and Systems Engineering

In Statistics and Probability

Citations in the corpus

Interval Coded Scoring: a toolbox for interpretable scoring systems (2018, methodology): “The use of regression as a CDSS has largely been restricted to logistic regression for likelihood estimation (Steyerberg et al., 2001).”
Constrain to perform: Regularization of habitat models (2006, background): “However, variable ection may actually reduce model performance, a as been suggested not to use it at all in small data Steyerberg et al., 2001).”
Prediction of Outcome in Traumatic Brain Injury with Computed Tomographic Characteristics: A Comparison between the Computed Tomographic Classification and Combinations of Computed Tomographic Predictors (2005, result): “This validation approach leads to better predictions of outcome for patients similar to the development population (13, 14, 30).”
Towards better clinical prediction models: seven steps for development and an ABCD for validation. (2014, methodology): “The GUSTO-I data set had 2851 deaths, which allows for reliable prediction modeling, where a common rule of thumb is to require at least ten events per variable (EPV) (2) (12).”
The number of subjects per variable required in linear regression analyses. (2015, background): “Steyerberg suggested three thresholds for the number of EPV in the context of accurate prediction of binary outcomes: 10 for any prediction modeling, 20 would remove the need for shrinkage of estimated regression coefficients in prespecified models, and 50 EPV would be required to permit reliable va”
Regression Modeling Strategies: With Applications to Linear Models, Logistic Regression, and Survival Analysis (2001, background): “• Decide where to spend them • Spend them • General references: [127, 159, 82, 66]”
Secure Logistic Regression Based on Homomorphic Encryption: Design and Evaluation (2018, background): “Many medical decision-making systems rely on the logistic regression model [7-9].”
Critical Appraisal and Data Extraction for Systematic Reviews of Prediction Modelling Studies: The CHARMS Checklist (2014, background): “However, such tests have frequently been criticised because of the limited statistical power to assess poor calibration and being oversensitive in large samples [58,117,120,121].”
What You See May Not Be What You Get: A Brief, Nontechnical Introduction to Overfitting in Regression-Type Models (2004, methodology): “Steyerberg et al. ( 14 ) have shown that models using preshrunken estimates and a fixed set of predictors tend to be the most likely to replicate in new samples.”
Prognosis and prognostic research: Developing a prognostic model (2009, background): “However, predictors that are not significant in univariable analysis should not be excluded as candidates.”
Association between postoperative troponin levels and 30-day mortality among patients undergoing noncardiac surgery. (2012, neutral): “For all models, forced simultaneous entry (all candidate variables remained in the models) was used rather than automated stepwise selection because simulations demonstrate a higher risk of overfitting with the latter approach.(18,19) We assessed colinearity using the variance inflation factor that ”
Focus on : Contemporary Methods in Biostatistics ( I ) Regression Modeling Strategies (2017, methodology): “05 value used as the stopping rule improves the selection of important variables in small datasets.(1) Stepwise procedures have also been associated with icreased probability of finding at least one of the variables significant by chance (type I error), due to multiple testing and no error-level adj”