moa-class {mogsa}R Documentation

Class "moa"

Description

moa class object

Objects from the Class

Objects can be created by calls of the form new("moa", ...).

Slots

eig:

eigen values

tau:

The percentage of explained variance by each datasets sparately.

partial.eig:

matrix, rows indicate the partial eigenvalues from each data.

eig.vec:

a matrix, eigenvectors.

loading:

the coordinate of variables/features.

fac.scr:

factor score of observations.

partial.fs:

partial factor score.

ctr.obs:

contribution of each observation to the total factor score.

ctr.var:

contribution of each variables to the total variance.

ctr.tab:

contribution of each data to the total variance.

RV:

pairwise RV coefficients

w.row:

weight of rows

w.data:

weight of datasets

data:

the original input data

tab.dim:

the dimension of each input data

call:

call

Methods

plot

signature(x = "moa", y = "missing"): Argument "value" sould be one of "eig", "tau", "obs", "var" and "RV"

if value = "eig", the eigenvalue would be plotted as scree plot. The following arguments could be set:

type=1 - The type of plot to show eigenvalues. (type=1: the eigenvalue are plotted; type=2: partial eigenvalue shown as concatenated bars; type=3: partial eigenvalue shown as bars side by side; type=4: matplot view of eigenvales, lty need to be set; type=5; the two dimensional plot of partial eigenvalues, axes and pch need to be set in this case.)

axes=NULL - The axes selected to plot

n=NULL - Top n eigenvalues to be drawn

tol=1e-5 - The tolerance of eigenvalue, eigenvalues lower than this value will not be shown.

legend=NULL - legend to put, a character string as calling legend function

col=NULL - The color of partial eigenvalues from each data set

lty=1 - The line type used in the matplot, used when type =4

pch=NULL - the pch to draw 2D partial eigen plot, when type = 5 used

lg.x="topright" - The position of legend

lg.y=NULL - Poistion argument passed to function "legend"

... - other arguemnts passed to functions

if value = "tau", the same with eig, but in the eigenvalues are scaled to 1

if value = "obs", the observation space will be shown, the following argument could be set:

axes=1:2 - Which axes should be draw

type=1 - Which type, see below (for type=1: the center points draw; type=2: the separate factor scores linked by lines; ... will be passed to function "points")

data.pch=20 - the pch of dataset, if type=1, the first one is used

col=1 - the color of observations, recycled used by data.frame

label=FALSE - A logical indicates if labels should be shown

lg.x="topright" - Position of legend

lg.y=NULL - Position of legend

xlim=NULL - The x limit

ylim=NULL - The y limit

label.cex=1 - the cex of text

...

var - the separate gene view, layout can be specified

RV - the heatmap of RV coefficients

show

signature(x = "moa", y = "missing"): show "moa" object

Author(s)

Chen Meng

References

Herve Abdi, Lynne J. Williams, Domininique Valentin and Mohammed Bennani-Dosse. STATIS and DISTATIS: optimum multitable principal component analysis and three way metric multidimensional scaling. WIREs Comput Stat 2012. Volume 4, Issue 2, pages 124-167

Herve Abdi, Lynne J. Williams, Domininique Valentin. Multiple factor analysis: principal component analysis for multitable and multiblock data sets. WIREs Comput Stat 2013

Examples

    showClass("moa")
    # load("R/mogsa/data/NCI60_4arrays.rda")
    data(NCI60_4arrays)
    ana <- moa(NCI60_4arrays, proc.row = "center_ssq1", w.data = "inertia", statis = TRUE)

    plot(ana, value="eig")
    plot(ana, value="tau", type=2)

[Package mogsa version 1.28.0 Index]