caellipse: Algebraic elliptical confidence regions for symmetrical...

Description Usage Arguments Details Value Note Author(s) References

View source: R/caellipse.R

Description

It produces elliptical confidence regions when symmetrical or ordered symmetrical correspondence analysis is performed. This function allows the analyst to superimpose confidence ellipses onto a graphical display when the input parameter catype of the main function CAvariants is set to "CA", "SOCA" or "DOCA". It is called internally from the main plot function plot.CAvariants. It uses the function ellipse.

Usage

1
2
3
4
caellipse(Xtable, a1 = 1, a2 = 2, alpha = 0.05, cols = c(2, 4), M = 2, cex = 0.8, 
cex.lab = 0.8, mar = c(5, 4, 4, 2) + 0.1, prop = 0.8, Imass, Jmass, a, b, g, f, dmu, 
inertiapc,  plottype = "biplot", biptype = "row", pos = 2, arrow = TRUE, length = 0, 
ell = TRUE)

Arguments

Xtable

The two-way contingency table.

a1

The axis number of the horizontal axis.

a2

The axis number of the vertical axis.

alpha

The confidence level of the elliptical regions. By default, alpha = 0.05.

cols

The graphical parameter for setting the colours of the points in the graphical displays.

M

The number of axes used for constructing the confidence ellypses. By default, M = 2. Its maximum value is equal to the rank of the data matrix.

cex

The parameter for setting the size of the character labels for the points in a graphical display. By default, cex = 0.8.

cex.lab

The parameter for setting the size of character labels of axes in graphical displays. By default, cex.lab = 0.8.

mar

The parameter for setting the size of the plotting area. By default, mar = c(5, 4, 4, 2) + 0.1.

prop

The scaling parameter for specifying the limits of the plotting area. By default, prop = 1.

Imass

The weight matrix of the row variable.

Jmass

The weight matrix of the column variable.

a

The row standard coordinates or, in case of the ordered variants of CA, the row standard polynomial coordinates.

b

The column standard coordinates or, in case of the ordered variants of CA, the column standard polynomial coordinates.

f

The row principal coordinates (scaled by a constant, by default scaleplot = 1).

g

The column principal coordinates (scaled by a constant, by default scaleplot = 1).

dmu

The squared singular values, or principal inertia, of each axis.

inertiapc

The percentage of explained inertia along each of the axes.

pos

The parameter that specifies the position of label of each point in the graphical display. By default, pos = 2.

plottype

The type of graphical display to be constructed. By default, plottype = "biplot"; the alternative is plottype = "classic".

biptype

The parameter for specifying the type of biplot. By default, biptype = "row".

arrow

The parameter used for diplaying the arrows in a biplot. By default, arrow = TRUE.

length

The parameter used for setting the length of the arrow end in a biplot. By default, length = 0.

ell

The logical parameter used for displaying the confidence ellipses. By default, ell = TRUE.

Details

The output values of this function.

Value

eccentricity

The eccentricity of the ellipses. This is the distance between the centre of the ellipse and its two foci, which can be thought of as a measure of how much the conic section deviates from being circular (when the region is perfectly circular, eccentricity is zero).

HL Axis 1

Value of the semi-major axis length for each row and column point.

HL Axis 2

Value of the semi-minor axis length for each row and column point.

Area

Area of the ellipse for each row and column point.

pvalcol

Approximate p-value for each of the row and column points.

Note

This function is called from the main plot function plot.CAvariants and is executed when ell = TRUE.

Author(s)

Rosaria Lombardo and Eric J Beh

References

Beh EJ 2010 Elliptical confidence regions for simple correspondence analysis. J. Stat. Plan. Inference 140, 2582–2588.
Beh EJ and Lombardo R 2014 Correspondence Analysis: Theory, Practice and New Strategies. Wiley.
Beh EJ Lombardo R 2015 Confidence regions and Approximate P-values for classical and non-symmetric correspondence analysis. Journal of Communications and Statistics, Theory and Methods. 44: 95–114.
Lombardo R Beh EJ 2016 Variants of Simple Correspondence Analysis. The R Journal, 8 (2), 167–184.


CAvariants documentation built on Jan. 14, 2019, 5:04 p.m.