Frequentist versus Bayesian approaches for AUC Confidence Interval Bounds - Inria - Institut national de recherche en sciences et technologies du numérique Accéder directement au contenu
Communication Dans Un Congrès Année : 2010

Frequentist versus Bayesian approaches for AUC Confidence Interval Bounds

Résumé

In this paper we first present two approaches, Frequen-tist and Bayesian, to calculate the Confidence Interval (CI) of Area Under the Curve (AUC). The goal of this studyis to compare both approaches and find out if they reveal significant differences along the sample size. We first generate a large number of hypothetical cases, based on True Negative (TN), True Positive (TP), False Positive (FP) and False Negative (FN), that lead to specific AUC values (90, 85, 80, 75,etc.). We then use both Frequentist and Bayesian approach to calculate the AUC CI bounds, AUCL and AUCH, and plot them for visual comparison. Results indicate that 1) for one sample size value the Bayesian approach can have multiple AUC CI bounds values, while the Frequentist has unique set of bounds, 2) for all sample size, the AUCL and AUCU values using the Frequentist approach are consistently under-estimated compared to the Bayesian ones, and 3) for very large sample size both approaches converge toward same values.
Fichier principal
Vignette du fichier
BHC-AUC-ISSPA.2010.pdf (927.97 Ko) Télécharger le fichier
Origine : Accord explicite pour ce dépôt
Loading...

Dates et versions

inria-00511188 , version 1 (23-10-2010)

Identifiants

  • HAL Id : inria-00511188 , version 1

Citer

Brahim Hamadicharef. Frequentist versus Bayesian approaches for AUC Confidence Interval Bounds. 10th International Conference on Information Science, Signal Processing and their Applications, Faculty of Biomedical & Health Science Engineering and Faculty of Electrical Engineering, Universiti Teknologi Malaysia (UTM), May 2010, Kuala Lumpur, Malaysia. pp.341-344. ⟨inria-00511188⟩

Collections

INSMI
85 Consultations
308 Téléchargements

Partager

Gmail Facebook X LinkedIn More