Partial Least Squares for compositional data applied to soil texture grape composition.

Gallo, Michele - Leone, Antonio - Amenta, Roberto (2010) Partial Least Squares for compositional data applied to soil texture grape composition. In: Agrostat2010: 11th European Symposium on Statistical Methods for the Food Industry., 24-26 Febbraio 2010, Benevento.

Documento PDF - Requires a PDF viewer such as GSview, Xpdf or Adobe Acrobat Reader
2321Kb

URL ufficiale: http://www.agrostat2010.unisannio.it/home.html

Abstract

Soil texture (i.e. the proportions of different-sized particles in a soil) may affect grape relationships mainly through influence on other soil properties, crucial to plant growth, such, for instance, water infiltration, hydraulic conductivity, water holding capacity. The sum of all soil particles is equal to 100%, thus texture can be considered as compositional data. Compositional data is commonly present in many disciplines. Nevertheless, too often the constant-sum constraint, which each compositional vector presents, is either ignored or improperly incorporated into statistical modeling and a misleading interpretation of the results is given. There are different approaches to incorporate compositional data into a statistical modeling, when it is not realistic to assume a multinomial distribution of the data. Following the approach proposed by Hinkle and Rayens (1995), in this paper we examine the problems that potentially occur when the PLS analysis (Wold, 1966) is performed on two set of compositional data. Up to now, it seems that this approach has never been used in studies pertaining the relationships between physical environmental variables and characteristics of the agricultural production, although many of the physical variables can be regarded as compositional. La texture du sol (c’est-à-dire la répartition dans le sol des minéraux par catégorie de grosseur) peut influencer la composition des raisins, principalement grâce à son influence sur d'autres propriétés importantes, qui sont cruciaux pour la croissance des plantes, telles que l'infiltration d'eau, la conductivité hydraulique, la capacité à retenir l'eau. La somme des particules minérales est égal à 100%, donc la texture peut être considérée comme une variable compositionnel. À la suite de l’approche proposée par Hinkle et Rayens (1995), dans cet article, nous examinons les problèmes qui pourraient survenir lors de l’analyse PLS (Wold, 1966) est effectuée sur deux séries de données compositionnel. Jusqu’à maintenant, il semble que cette approche n’a jamais été utilisée dans les études concernant les relations entre les variables de l’environnement physique et les caractéristiques de la production agricole, bien que bon nombre des variables physiques peuvent être considérés comme compositionnel.

Tipologia del documento:Contributo a convegno o workshop (Comunicazione)
Parole chiave:Soil texture, Grape composition, Compositional Data, Logcontrast, PLS
Settori scientifico-disciplinari del MIUR:AREA 13 - Scienze economiche e statistiche > STATISTICA
Codice identificativo (ID):704
Depositato da:Prof. Michele Gallo
Depositato il:19 Ago 2011 09:32
Ultima modifica:19 Ago 2011 09:32

Riservato allo Staff dell'Archivio: Accedi al record del documento