Announce of Seminar
Prof. Peter Filzmoser
Department of Statistics and Probability Theory Vienna University of Technology
Compositional data analysis: Challenges for environmental sciences
13 June, 2012 at 15.30 Aula E, Santa Marta, Venezia
Anyone interested is invited to participate
Many practical data sets in environmental sciences are in fact compositional data because only the ratios between the variables are informative. Compositional data are represented in the Aitchison geometry on the simplex, and for applying statistical methods designed for the Euclidean geometry they need to be transformed first. The isometric logratio transformation has the best geometrical properties, and it avoids the singularity problem introduced by the centered logratio transformation. Since real environmental data also often contain outlying observations, robust statistical procedures need to be employed. We show for different multivariate methods how robustness can be managed for compositional data, provide algorithms for the computation, and apply the methods on real data sets from geochemistry.
The announcement is available here.