Taux d'erreur; Classement erroné; Fonction discriminante linéaire; Bootstrap; Jackknife; Validation croisée; Monte Carlo; Error rates; Misclassification; Linear discriminant function; Crossvalidation
Abstract :
[fr] A partir de simulations, on compare les performances de dix estimateurs paramétriques et six estimateurs non paramétriques dans le cas de l'estimation de trois taux d'erreur en analyse discriminante (taux d'erreur réel, attendu et optimal). Cette comparaison est limitée au cas de l'analyse discriminante linéaire appliquée à deux populations normales de même matrice de variances et covariances. Les simulations ont montré qu'il existe des différences importantes entre les estimateurs et l'utilisateur doit être mis en garde vis-à-vis de certains estimateurs proposés dans les logiciels statistiques. [en] Monte Carlo experiments are performed to compare ten parametric and six non parametric estimators of the three error rates (actual, expected and optimal error rate) in discriminant analysis. This comparison is restricted to the context of two p-dimensional normal group conditional distributions with the same covariance matrix. The study shows that certain estimators are markedly superior and users should be warned against some estimators available in statistical softwares.
Disciplines :
Mathematics Life sciences: Multidisciplinary, general & others
Author, co-author :
Piraux, François; Faculté Universitaire des Sciences Agronomiques de Gembloux - FUSAGx > Sciences agronomiques > Statistique, informatique et mathématique appliquées
Palm, Rodolphe ; Faculté Universitaire des Sciences Agronomiques de Gembloux - FUSAGx > Sciences agronomiques > Statistique, Inform. et Mathém. appliquées
Language :
French
Title :
Etude empirique des estimateurs des taux d'erreur en analyse discriminante
This website uses cookies to improve user experience. Read more
Save & Close
Accept all
Decline all
Show detailsHide details
Cookie declaration
About cookies
Strictly necessary
Performance
Strictly necessary cookies allow core website functionality such as user login and account management. The website cannot be used properly without strictly necessary cookies.
This cookie is used by Cookie-Script.com service to remember visitor cookie consent preferences. It is necessary for Cookie-Script.com cookie banner to work properly.
Performance cookies are used to see how visitors use the website, eg. analytics cookies. Those cookies cannot be used to directly identify a certain visitor.
Used to store the attribution information, the referrer initially used to visit the website
Cookies are small text files that are placed on your computer by websites that you visit. Websites use cookies to help users navigate efficiently and perform certain functions. Cookies that are required for the website to operate properly are allowed to be set without your permission. All other cookies need to be approved before they can be set in the browser.
You can change your consent to cookie usage at any time on our Privacy Policy page.