Step 9 – Analysis of outliers

NOTE : In this chart, the cloud points provide far more information than the one presented in the unprocessed datable with outliers present (-> https://github.com/MoiraCorp/DLMM-IRating-in-R/tree/main/steps/step8/alternr)

In order to compute the matrix of p-value, we use the custom cor.pvalue() R function already introduced in chapter “step8”: -> https://github.com/MoiraCorp/DLMM-IRating-in-R/tree/main/steps/step8/selectvar

# Function computing the matrix of p-values
# mat : is a matrix of data
# … : further arguments to pass to the native R cor.test function
cor.pvalue <- function(mat, …) {
mat <- as.matrix(mat)
n <- ncol(mat)
p.mat<- matrix(NA, n, n)
diag(p.mat) <- 0
for (i in 1:(n – 1)) {
for (j in (i + 1):n) {
tmp <- cor.test(mat[, i], mat[, j], …)
p.mat[i, j] <- p.mat[j, i] <- tmp$p.value
}
}
colnames(p.mat) <- rownames(p.mat) <- colnames(mat)
p.mat
}

In order to produce a graphics representation of the Pearson correlation between all NA masqued Ratio Variables we are using the corrplot R package: -> https://cran.r-project.org/web/packages/corrplot/vignettes/corrplot-intro.html

library(corrplot)
corrprs <- cor(wcs2train.ratios.NA, use=”pairwise”, method=”pearson”)
p.mat <- cor.pvalue(wcs2train.ratios.NA)
col <- colorRampPalette(c(“#BB4444”, “#fcc3b8”, “#FFFFFF”, “#add2f7”, “#4fc69d”))
corrplot(corrprs, method=”color”, col=col(200),
type=”upper”,
addCoef.col = “black”, # Add coefficient of correlation
addCoefasPercent = TRUE,
tl.col=”black”, tl.srt=45, #Text label color and rotation
# Combine with significance
p.mat = p.mat, sig.level = 0.01, insig = “blank”,
# hide correlation coefficient on the principal diagonal
diag=FALSE
)

The graphics representation of the Pearson correlation between all Na masqued Ratio Variables is presented in Table_4_21c_Page 164_RatioswithNA_Correlation.pdf