Statistique de rangs et Analyse statistique implicative

Abstract : This communication explores the benefit of one data analysis method implicative
analysis such as Gras defined it to study ranks concordance-discordance accorded by judges
to things. This meaning is the same as in Friedman's or Kendall's works. Then we compare
one analysis of judges' preferences through ranks to the propensity analysis between modal
variables built by Lagrange. To achieve it, we don't consider the a priori independence
hypothesis between variables. Nevertheless, we give a quality measure to "theorem" : "if an
A thing is ranked by judges, then generally a B thing is ranked in a better way by the same
judges" and we represent by graph relations of preferences for the whole ranked objects. We
limit this study to the two cases of complete ranking and incomplete ranking, without ex
æquo between q things by k judges.
Complete list of metadatas
Contributor : Jean-Claude Regnier <>
Submitted on : Tuesday, June 16, 2009 - 6:17:19 PM
Last modification on : Friday, June 14, 2019 - 1:20:54 AM


  • HAL Id : halshs-00396031, version 1


Jean-Claude Regnier, Régis Gras. Statistique de rangs et Analyse statistique implicative. Régis Gras; Jean-Claude Régnier; Fabrice Guillet. Analyse statistique implicative : Une méthode d'analyse de données pour la recherche de causalités, Éditions Cépaduès, pp.131-149, 2009, RNTI-E-16. ⟨halshs-00396031⟩



Record views