Interval-Valued Linear Model - Archive ouverte HAL Accéder directement au contenu
Article Dans Une Revue International Journal of Computational Intelligence Systems Année : 2015

Interval-Valued Linear Model

Résumé

This paper introduces a new type of statistical model: the interval-valued linear model, which describes the linear relationship between an interval-valued output random variable and real-valued input variables. Firstly, notions of variance and covariance of set-valued and interval-valued random variables are introduced. Then, we give the definition of the interval-valued linear model and its least square estimation, as well as some properties of the least square estimator (LSE). Thirdly, we show that, whereas the best linear unbiased estimation does not exist, the best binary linear unbiased estimator exists and it is the LSE. Finally, we present simulation experiments and an application example regarding temperatures of cities affected by their latitude, which illustrates the application of the proposed model.
Fichier principal
Vignette du fichier
interval-valued linear model.pdf (510.59 Ko) Télécharger le fichier
Origine : Fichiers produits par l'(les) auteur(s)
Loading...

Dates et versions

hal-01127794 , version 1 (08-03-2015)

Identifiants

Citer

Xun Wang, Shoumei Li, Thierry Denoeux. Interval-Valued Linear Model. International Journal of Computational Intelligence Systems, 2015, 8 (1), pp.114-127. ⟨10.1080/18756891.2014.967010⟩. ⟨hal-01127794⟩
100 Consultations
157 Téléchargements

Altmetric

Partager

Gmail Facebook X LinkedIn More