correlationSelection: correlationSelection: A vector correlation function...

In bertamiro/lpattern: Detection of scatterplots with L-shaped pattern

Description

correlationSelection Uses the function matCorrs; given two matrices X (m,n) , Y (m,n) this function computes Pearson and Spearman correlation coefficients and their significance p-values for every pair of row vectors.

Usage

correlationSelection(X, Y, type = "Spearman", adj = TRUE,
  pValCutoff = 0.05, rCutoff = 0, sortByCorrs = FALSE)

Arguments

X	First matrix
Y	Second matrix. Must have the same dimensions as X.
type	specifies the correlation to choose between Spearman and Pearson. Default is Spearman.
adj	logical variable indicating if the p-value returned should be adjusted or not. Default set to TRUE, which will return an adjusted p-value.
pValCutoff	the upper limit to be used for the p-value. Default is 0.05.
rCutoff	the upper limit to be used for the correlation coefficient. Default is 0, no cut off.
sortByCorrs	logical; if TRUE, results will be ordered in ascending order by p-value. Default set to FALSE.

Examples

# (X <- round(matrix (rnorm(30)*10, ncol=6),1))
# (Y <- round(X + matrix (rnorm(30)*10, ncol=6),1))
# (rownames(X)=rownames(Y)=letters[1:nrow(X)])
# (m1<-matCorrs(X,Y))
# (m2<-matDistCorr(X,Y))
# (m12<- matAllCorrs (X, Y))
# correlationSelection(X, Y,pValCutoff=0.25)
# correlationSelection(X, Y, pValCutoff=0.25, type="Pearson")
# correlationSelection(X, Y, pValCutoff=0.25, rCutoff=0.1, type="Spearman", sortByCorrs=TRUE)
# sort1(m12,1)
# sort1(m12,3)
# dcor(X,Y,1)
# coeffRV(X,Y)
# multivCorr(X,Y)

#Adding an NA to X will make the preceeding fail
# X[1,1] <-NA
# (m1<-matCorrs(X,Y))
# (m2<-matDistCorr(X,Y))
# (m12<- matAllCorrs (X, Y))
# sort1(m12,4, DEC=FALSE)

bertamiro/lpattern documentation built on July 19, 2019, 12:56 p.m.