coefcorr: Subspatial Correlation Coefficients

In victorvicpal/MGSR: Multivariate Gaussian Subspatial Regression

Description

Subspatial Correlation Coefficients calculation for each subpatial function (lmc fitting).

Usage

coefcorr(RES)

Arguments

RES	Resulting lmc fitting.

Details

It allows for knowing how subspatially related are the variables.

Value

Object containing the subspatial correlation matrices.

Author(s)

Victor Vicente Palacios

See Also

lmc

Examples

data("iris")
Versicolor <- iris[which(iris$Species=='versicolor'),-5]
##Data Standarization
means_vers <- apply(Versicolor,2,mean)
sd_vers <- apply(Versicolor,2,sd)
Versicolor_st <- Versicolor

for (i in 1:length(Versicolor[1,]))
{Versicolor_st[,i] <- (Versicolor[,i]-means_vers[i])/sd_vers[i]}

##PrComp
PC_train <- princomp(Versicolor_st)
## CrossVariogram Calculation
CV_vers <- crossvariogram(as.data.frame(PC_train$scores[,1:2]),as.data.frame(Versicolor_st),11)
## lmc 
RES_vers <- lmc(CV_vers,'Pow',1.6)
## subspatial correlation coefficients
sub_coef <- coefcorr(RES_vers)

victorvicpal/MGSR documentation built on May 3, 2019, 6:11 p.m.