common components and specific weights analysis, algorithmm different from ccswa
result= comdim(col,lv,(threshold));
a vector (ntab x 1) of Div structures containing each a table;
all the tables have the same number of observations n
the table i has pi variables
number of components; dimension max. of the model
threshold for the convergence; default=10^(-10)
scores of the observations on the common componants
result.scores.d is a matrix of dimensions (n x lv)
weight of each table according to the number of components
result.weights.d is a vector of dimensions (ntab x lv)
scores of each table
result.scores_tables.d is a vector of dimensions (lv x 1)