DiStatis-DiStatis: 'DiStatis' of a DiStatis object High level constructor of...

Description Usage Arguments Format Details Value Note Author(s) References Examples


This is the function that makes DiStatis Methodology: Statis is part of the PCA family and therefore the main analytical tool for STATIS is the singular value decomposition (SVD) and the generalized singular value decomposition (GSVD) of a matrix. The goal of Statis is to analyze several data sets of variables that were collected on the same set of observations. Originally, the comparisons were drawn from the compute of the scalar product between the different tables. In this approach, the condition is made more flexible, allowing the incorporation of different distance measurements (including the scalar product) to compare the tables.

function to do Statis (K-tables methodology)with distance options


DiStatis(Data = NULL, Distance = c(), Center = TRUE, Scale = TRUE,
  CorrelVector = TRUE, Frec = FALSE, Traj = TRUE)



The data frame or of k-tables type. The Observations should be in rows (common elements in DAnisostatis), the variables and Studies must be in columns. After the name of the variable an underscore (_) must be written to indicate the Study (eg. Var1_Est1 , eg. Var1_EstK, for more information see the data object). The name of a variable can include any symbol except an underscore (_). REMEMBER the underscore (_) should be reserved to indicate the study. Also, the Data can be a list of k components. Each element of the list is one of the tables with observations in rows and variables in columns. The elements of list must be data.frame or ExpressionSet data.


Vector is the length equal to the number of studies that indicates the kind of distance (or scalar product) that is calculated in each study. If not specify (or is wrong specify) the scalar product is used. The options can be ScalarProduct, euclidean, manhattan, canberra, pearson, pearsonabs, spearman, spearmanabs, mahalanobis. In the binary data the distance can be: jaccard, simple matching, sokal&Sneath, Roger&Tanimoto, Dice, Hamman,#' Ochiai, Sokal&Sneath, Phi-Pearson, Gower&Legendre.


A logical value. If TRUE, the data frame is centered by the mean. By default is TRUE.


A logical value indicating whether the column vectors (of the data.frame) should be standardized by the rows weight, by default is TRUE.


a logical value. If TRUE (default), Vectorial correlation coefficient is computed for the RV matrix. If FALSE the Hilbert-Smith distance is used in the RV matrix.


Logical. Should the data be treated data as frequencies? By default is FALSE.


Logical. Should the trajectories analysis be done? By default is TRUE.


An object of class NULL of length 0.


STATIS methods: to more information, see references.



DiStatis class object with the corresponding completed slots according to the given model


use DiStatis-class high level constructor for the creation of the class instead of directly calling its constructor by new means.


M L Zingaretti, J A Demey-Zambrano, J L Vicente Villardon, J R Demey


  1. Abdi, H., Williams, L.J., Valentin, D., & Bennani-Dosse, M. (2012). STATIS and DISTATIS: optimum multitable principal component analysis and three way metric multidimensional scaling. WIREs Comput Stat, 4, 124-167.

  2. Escoufier, Y. (1976). Operateur associe a un tableau de donnees. Annales de laInsee, 22-23, 165-178.

  3. Escoufier, Y. (1987). The duality diagram: a means for better practical applications. En P. Legendre & L. Legendre (Eds.), Developments in Numerical Ecology, pp. 139-156, NATO Advanced Institute, Serie G. Berlin: Springer.

  4. L'Hermier des Plantes, H. (1976). Structuration des Tableaux a Trois Indices de la Statistique. [These de Troisieme Cycle]. University of Montpellier, France.



kimod documentation built on April 28, 2020, 9:21 p.m.