common components and co-inertia analysis
result= acom1(col,lv);
a vector (ntab x 1) of Div structures, each of them containing a matrix i, i=1:ntab;
all the matrices have the same number of observations n
the table i has pi variables
number of components; the dimension max of the model
global scores
result.global_scores.d is a matrix of dimensions (n x lv)
global loadings
result.global_loadings.d is a matrix of dimensions ((sum of pi) x lv)
scores of each table i; a vector of Div structures of dimensions (ntab x 1)
result.individual_scores(i).d is a matrix of dimensions (n x lv)
scores of each component; a vector of Div structures of dimensions (ntab x 1)
result.individual_loadings(i).d is a matrix of dimensions (pi x lv)
Frobenius norm of each table; a Div structure
result.individual_tablenorm.d is a vector of length ntab
vectors of the means of each table; a vector of Div structures of dimensions (ntab x 1)
result.individual_mean(i).d is a vector of dimensions (pi x 1)
concatened matrix of individual scores, useful for barycenter representation; a Div structures of dimensions ((n x ntab) x lv)
the number of variables of each table; a Div structure
result.tables_size.d is a vector of dimensions (ntab x 1), containing the pi
the scores associated to each table; a Div structure
result.tables_scores.d is a matrix of dimensions (ntab x lv), containing the pi
correlations
result.tables_scores.d is a matrix of dimensions (ntab x lv)
result.explained_sum_squares.d is a matrix of dimensions (ntab x lv)