ade4 (version 1.5-2)

wca: Within-Class Analysis

Description

Performs a particular case of an Orthogonal Principal Component Analysis with respect to Instrumental Variables (orthopcaiv), in which there is only a single factor as covariable.

Usage

within(dudi, fac, scannf = TRUE, nf = 2)
## S3 method for class 'dudi':
wca(x, fac, scannf = TRUE, nf = 2, \dots)

Arguments

dudi

a duality diagram, object of class dudi obtained from the functions dudi.coa, dudi.pca,...

x

a duality diagram, object of class dudi from one of the functions dudi.coa, dudi.pca,...

fac

a factor partitioning the rows of dudi$tab in classes

scannf

a logical value indicating whether the eigenvalues bar plot should be displayed

nf

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

...

further arguments passed to or from other methods

Value

• Returns a list of the sub-class within in the class dudi
• tab
  a data frame containing the transformed data (substraction of the class mean)
• call
  the matching call
• nf
  number of kept axes
• rank
  the rank of the analysis
• ratio
  percentage of within-class inertia
• eig
  a numeric vector containing the eigenvalues
• lw
  a numeric vector of row weigths
• cw
  a numeric vector of column weigths
• tabw
  a numeric vector of class weigths
• fac
  the factor defining the classes
• li
  data frame row coordinates
• l1
  data frame row normed scores
• co
  data frame column coordinates
• c1
  data frame column normed scores
• ls
  data frame supplementary row coordinates
• as
  data frame inertia axis onto within axis

encoding

latin1

References

Benz�cri, J. P. (1983) Analyse de l'inertie intra-classe par l'analyse d'un tableau de correspondances. Les Cahiers de l'Analyse des donn�es, 8, 351--358. Dol�dec, S. and Chessel, D. (1987) Rythmes saisonniers et composantes stationnelles en milieu aquatique I- Description d'un plan d'observations complet par projection de variables. Acta Oecologica, Oecologia Generalis, 8, 3, 403--426.

Examples

data(meaudret)
par(mfrow = c(2,2))
pca1 <- dudi.pca(meaudret$env, scan = FALSE, nf = 4)
s.traject(pca1$li, meaudret$design$site, 
    sub = "Principal Component Analysis", csub = 1.5)
wit1 <- wca(pca1, meaudret$design$site, scan = FALSE, nf = 2)
s.traject(wit1$li, meaudret$design$site, 
    sub = "Within site Principal Component Analysis", csub = 1.5)
s.corcircle (wit1$as)
par(mfrow = c(1,1))
plot(wit1)