sepan function

Separated Analyses in a K-tables

performs K separated multivariate analyses of an object of class ktab

containing K tables.

sepan(X, nf = 2)
## S3 method for class 'sepan'
plot(x, mfrow = NULL, csub = 2, ...)
## S3 method for class 'sepan'
summary(object, ...)
## S3 method for class 'sepan'
print(x, ...)

Arguments

• X: an object of class ktab
• nf: an integer indicating the number of kept axes for each separated analysis
• x, object: an object of class 'sepan'
• mfrow: a vector of the form "c(nr,nc)", otherwise computed by a special own function n2mfrow
• csub: a character size for the sub-titles, used with par("cex")*csub
• ...: further arguments passed to or from other methods

Returns

returns a list of class 'sepan' containing : - call: a call order

• tab.names: a vector of characters with the names of tables

• blo: a numeric vector with the numbers of columns for each table

• rank: a numeric vector with the rank of the studied matrix for each table

• Eig: a numeric vector with all the eigenvalues

• Li: a data frame with the row coordinates

• L1: a data frame with the row normed scores

• Co: a data frame with the column coordinates

• C1: a data frame with the column normed coordinates

• TL: a data frame with the factors for Li L1

• TC: a data frame with the factors for Co C1

Details

The function plot on a sepan object allows to compare inertias and structures between arrays. In black, the eigenvalues of kept axes in the object 'sepan'.

Author(s)

Daniel Chessel

Examples

data(escopage)
w <- data.frame(scale(escopage$tab))
w <- ktab.data.frame(w, escopage$blo, tabnames = escopage$tab.names)
sep1 <- sepan(w)
sep1
summary(sep1)
plot(sep1)