ade4 (version 1.01)

statis: STATIS, a method for analysing K-tables

Description

performs a STATIS analysis of K-tables.

Usage

statis(X, scannf = TRUE, nf = 3, tol = 1e-07)
plot.statis (x, xax = 1, yax = 2, option = 1:4, ...) 
print.statis (x, ...)

Arguments

X

an object of class 'ktab'

scannf

a logical value indicating whether the number of kept axes for the compromise should be asked

nf

if scannf FALSE, an integer indicating the number of kept axes for the compromise

tol

a tolerance threshold to test whether the distance matrix is Euclidean : an eigenvalue is considered positive if it is larger than -tol*lambda1 where lambda1 is the largest eigenvalue

statis, x

an object of class 'statis'

xax, yax

the numbers of the x-axis and the y-axis

option

an integer between 1 and 4, otherwise the 4 components of the plot are dispayed

...

further arguments passed to or from other methods

Value

• statis returns a list of class 'statis' containing :
• $RV
  a matrix with the all RV coefficients
• RV.eig
  a numeric vector with all the eigenvalues
• RV.coo
  a data frame with the array scores
• tab.names
  a vector of characters with the names of the arrays
• $RV.tabw
  a numeric vector with the array weigths
• nf
  an integer indicating the number of kept axes
• rank
  an integer indicating the rank of the studied matrix
• C.li
  a data frame with the row coordinates
• C.Co
  a data frame with the column coordinates
• T4
  a data frame with the principal vectors (for each table)
• TL
  a data frame with the factors (not used)
• TC
  a data frame with the factors for Co
• T4
  a data frame with the factors for T4

References

Lavit, C. (1988) Analyse conjointe de tableaux quantitatifs, Masson, Paris. Lavit, C., Escoufier, Y., Sabatier, R. and Traissac, P. (1994) The ACT (Statis method). Computational Statistics and Data Analysis, 18, 97--119.

Examples

data(jv73)
kta1 <- ktab.within(within.pca(jv73$morpho, jv73$fac.riv, scann = FALSE))
statis1 <- statis(kta1, scann = FALSE)
plot(statis1)

dudi1 <- dudi.pca(jv73$poi, scann = FALSE, scal = FALSE)
wit1 <- within(dudi1, jv73$fac.riv, scann = FALSE)
kta3 <- ktab.within(wit1)
data(jv73)
statis3 <- statis(kta3, scann = FALSE)
plot(statis3)

s.arrow(statis3$C.li, cgrid = 0)

kplot(statis3, traj = TRUE, arrow = FALSE, unique = TRUE, 
    clab = 0, csub = 3, cpoi = 3)
statis3