ATTENTION/WARNING - NE PAS DÉPOSER ICI/DO NOT SUBMIT HERE

Ceci est la version de TEST de DIAL.mem. Veuillez ne pas soumettre votre mémoire sur ce site mais bien à l'URL suivante: 'https://thesis.dial.uclouvain.be'.
This is the TEST version of DIAL.mem. Please use the following URL to submit your master thesis: 'https://thesis.dial.uclouvain.be'.
 

Receiver Operating Characteristic Curves

Ourrad, Adam
(2022)

Files

Ourrad_82391900_2022.pdf
  • Closed access
  • Adobe PDF
  • 2.07 MB

Details

Supervisors
El Ghouch, Anouar
Faculty
Faculté des sciences
Degree label
Master [120] en statistique, orientation générale, à finalité spécialisée
Abstract
In a logistic regression model, to make a diagnostic or a decision, classifying the probability estimate that the output belongs to a group (say 0 for no disease and 1 for disease) is necessary to fix a discriminant threshold (or optimal cutoff point). The choice of such a classifier is crucial to correctly estimate the value of the output and different methods exist to provide it but this system is imperfect. That is why Receiver Operating Characteristic curve (ROC curve) is a graphical (and analytical) tool to find out this optimal value of this classifier using different performance measures criteria and to sum up classifier performance. It can also be critical to make a choice between these different available classifiers (cutoff points). In this study, different methods to obtain such an optimal threshold will be seen, and the comparison between them will be effectuated to find out the value that will best estimate the output.