Pharmacokinetic Models: Comparison and Selection Criterion
Decision Making: P-value Method
Quantifying and Rejecting Outliers: The Grubbs Test
Survival Tree
Frequency-dependent Selection
Variability: Analysis
You might also read
Articles linked to this work by shared authors, journal, and citation graph.
Updated: Jun 24, 2025

Selecting Multiple Biomarker Subsets with Similarly Effective Binary Classification Performances
Published on: October 11, 2018
Arnab Kumar Maity1, Sanjib Basu2, Santu Ghosh3
1Pfizer Inc., San Diego, CA.
The Deviance Information Criterion (DIC) often overfits models, showing high sensitivity but poor correct selection rates (0-2%) in Bayesian model selection, even with large sample sizes. Marginal likelihood criteria offer better asymptotic performance, avoiding DIC's persistent mis-selection issues.
Area of Science:
Background:
Purpose of the Study:
Main Methods:
Main Results:
Conclusions: