Geodesic
Covariance Structure of Principal Components for Three-Part Compositional Data
Acta Universitatis Palackianae Olomucensis. Facultas Rerum Naturalium. Mathematica, Tome 52 (2013) no. 2, pp. 61-69 Cet article a éte moissonné depuis la source Czech Digital Mathematics Library

Voir la notice de l'article

Statistical analysis of compositional data, multivariate observations carrying only relative information (proportions, percentages), should be performed only in orthonormal coordinates with respect to the Aitchison geometry on the simplex. In case of three-part compositions it is possible to decompose the covariance structure of the well-known principal components using variances of log-ratios of the original parts. They seem to be helpful for the interpretation of these special orthonormal coordinates. Theoretical results are applied to real-world data containing relative structure of landscape use in German regions.
Statistical analysis of compositional data, multivariate observations carrying only relative information (proportions, percentages), should be performed only in orthonormal coordinates with respect to the Aitchison geometry on the simplex. In case of three-part compositions it is possible to decompose the covariance structure of the well-known principal components using variances of log-ratios of the original parts. They seem to be helpful for the interpretation of these special orthonormal coordinates. Theoretical results are applied to real-world data containing relative structure of landscape use in German regions.
Classification : 15A18, 62H25, 62H99, 62J10
Keywords: compositional data; covariance structure; principal components; log-contrasts 
@article{AUPO_2013_52_2_a5,
     author = {Hr\r{u}zov\'a, Kl\'ara and Hron, Karel and Rypka, Miroslav and Fi\v{s}erov\'a, Eva},
     title = {Covariance {Structure} of {Principal} {Components} for {Three-Part} {Compositional} {Data}},
     journal = {Acta Universitatis Palackianae Olomucensis. Facultas Rerum Naturalium. Mathematica},
     pages = {61--69},
     year = {2013},
     volume = {52},
     number = {2},
     mrnumber = {3202380},
     zbl = {06296015},
     language = {en},
     url = {http://geodesic.mathdoc.fr/item/AUPO_2013_52_2_a5/}
}
TY  - JOUR
AU  - Hrůzová, Klára
AU  - Hron, Karel
AU  - Rypka, Miroslav
AU  - Fišerová, Eva
TI  - Covariance Structure of Principal Components for Three-Part Compositional Data
JO  - Acta Universitatis Palackianae Olomucensis. Facultas Rerum Naturalium. Mathematica
PY  - 2013
SP  - 61
EP  - 69
VL  - 52
IS  - 2
UR  - http://geodesic.mathdoc.fr/item/AUPO_2013_52_2_a5/
LA  - en
ID  - AUPO_2013_52_2_a5
ER  - 
%0 Journal Article
%A Hrůzová, Klára
%A Hron, Karel
%A Rypka, Miroslav
%A Fišerová, Eva
%T Covariance Structure of Principal Components for Three-Part Compositional Data
%J Acta Universitatis Palackianae Olomucensis. Facultas Rerum Naturalium. Mathematica
%D 2013
%P 61-69
%V 52
%N 2
%U http://geodesic.mathdoc.fr/item/AUPO_2013_52_2_a5/
%G en
%F AUPO_2013_52_2_a5
Hrůzová, Klára; Hron, Karel; Rypka, Miroslav; Fišerová, Eva. Covariance Structure of Principal Components for Three-Part Compositional Data. Acta Universitatis Palackianae Olomucensis. Facultas Rerum Naturalium. Mathematica, Tome 52 (2013) no. 2, pp. 61-69. http://geodesic.mathdoc.fr/item/AUPO_2013_52_2_a5/

[1] Aitchison, J.: The Statistical Analysis of Compositional Data. Chapman and Hall, London, 1986. | MR | Zbl

[2] Egozcue, J. J., Pawlowsky-Glahn, V., Mateu-Figueras, G., Barceló-Vidal, C.: Isometric logratio transformations for compositional data analysis. Math. Geol. 35 (2003), 279–300. | DOI | MR

[3] Egozcue, J. J., Pawlowsky-Glahn, V.: Groups of parts and their balances in compositional data analysis. Math. Geol. 37 (2005), 795–828. | DOI | MR | Zbl

[4] Filzmoser, P., Hron, K.: Robustness for compositional data. In: Becker, C., Fried, R., Kuhnt, S. (eds.): Robustness and complex data structures, Springer, Heidelberg, 2013, 117–131. | MR

[5] Fišerová, E., Hron, K.: On the interpretation of orthonormal coordinates for compositional data. Math. Geosci. 43 (2011), 455–468. | DOI

[6] Fišerová, E., Hron, K.: Statistical inference in orthogonal regression for three-part compositional data using a linear model with type-II constraints. Communications in Statistics – Theory and Methods 41 (2012), 2367–2385. | MR | Zbl

[7] Golub, G. H.: Modified matrix eigenvalue problems. SIAM Review 15 (1973), 318–334. | DOI | MR | Zbl

[8] Pawlowsky-Glahn, V., Buccianti, A.: Compositional data analysis: Theory and applications. Wiley, Chichester, 2011. | MR

[9] R Development Core Team: R: A language and environment for statistical computing. R Foundation for Statistical Computing, Vienna, 2013.

[10] Härdle, W. K., Simar, L.: Applied Multivariate Statistical Analysis. Springer-Verlag, Berlin, Heidelberg, 2012. | MR | Zbl

[11] Jackson, J. D., Dunlevy, J. A.: Orthogonal least squares and the interchangeability of alternative proxy variables in the social sciences. J. Roy. Statist. Soc. Ser. D (The Statistician) 37, 1 (1988), 7–14.

[12] Landesbetrieb für Statistik und Kommunikationstechnologie Niedersachsen: Kreiszahlen: Ausgewählte Regional Daten für Deutschland. Landesbetrieb für Statistik und Kommunikationstechnologie Niedersachsen, Hannover, 2012.