Search code examples
rcorrelationdiagonalr-corrplot

How to set R corrplot diagonal numeric labels?

I want to get labels on diagonal like in Fig. 3 but with corrplot like in Fig. 1-2. I am studying corrplot manual here for the numeric diagonal labels. I do not know any method which would allow to put numeric labels on corrplot diagonal because I have managed to falsify all potential choices. Falsified things

  • numeric diagonal labels cannot be set by the following line in the function cor.mtest

    colnames(p.mat) <- rownames(p.mat) <- colnames(mat) <- diag.labels

  • colorlegend is apparently not the right choice here

    corrplot(...)
colorlegend(colbar = grey(1:100 / 100), labels=ids, addlabels = TRUE)

Partial things nice to know more but not limiting us

  • diag=FALSE, tl.pos="d" is for a single cell. How can you make tl.pos="d" for N amount of cells? - - The tl.pos=c("d") will lead to an error. - - Is tl.pos necessary for the numerical diagonal labels?

Code but also here for different examples where included K.J.J.K's first answer's proposal as a test case but which is shown to be false for the task

library("corrplot")

# http://www.sthda.com/english/wiki/visualize-correlation-matrix-using-correlogram
cor.mtest <- function(mat, diag.labels, ...) {
    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) <- diag.labels
  p.mat
}

ids <- c(seq(1,11))

M<-cor(mtcars)
p.mat <- cor.mtest(mtcars, diag.labels=ids)
corrplot(M, type="upper", order="hclust", diag=FALSE, # TODO tl.pos=c("d"),
         p.mat = p.mat, sig.level = 0.05)

Fig. 1 Output where no expected labels on the diagonal, Fig. 2 Falsifying K.J.J.K's proposal where no effect on the diagonal labels, Fig. 3 Example of labels on diagonal with corrgram found here

enter image description here enter image description here enter image description here

Expected output: numeric labels on the diagonal like in Fig. 3 but decoration otherwise wanted like in Fig. (1-2)

Falsifying K.J.J.K's proposal

Get the code here and you get the output where no change on the diagonal labels in Fig. 2.

OS: Debian 8.5
R: 3.3.1
Ticket in Developer's Github: #71

Solution 
  • library("corrplot")

# http://rstudio-pubs-static.s3.amazonaws.com/6382_886fbab74fd5499ba455f11360f78de7.html
# plotcorr(R, col = colorRampPalette(c("#E08214", "white", "#8073AC"))(10), type = "lower")

# http://www.sthda.com/english/wiki/visualize-correlation-matrix-using-correlogram
# corrplot(M, type="upper", order="hclust", tl.col="black", tl.srt=45)

## Compute p-value of correlations
# mat : is a matrix of data
# ... : further arguments to pass to the native R cor.test function

M<-cor(mtcars)

# http://www.sthda.com/english/wiki/visualize-correlation-matrix-using-correlogram
cor.mtest <- 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
}
# matrix of the p-value of the correlation
p.mat <- cor.mtest(mtcars)
head(p.mat[, 1:5])

corrplot(M, type="upper", order="hclust", 
  p.mat = p.mat, sig.level = 0.05)

# Leave blank on no significant coefficient
corrplot(M, type="upper", order="hclust", 
  p.mat = p.mat, sig.level = 0.01, insig = "blank")

col <- colorRampPalette(c("#BB4444", "#EE9988", "#FFFFFF", "#77AADD", "#4477AA"))
corrplot(M, method="color", col=col(200),  
  type="upper", order="hclust", 
  addCoef.col = "black", # Add coefficient of correlation
  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 
)

ids <- c(seq(1,11))
M<-cor(mtcars)
colnames(M)<-ids
rownames(M)<-c("I","told","you","row","names","controls","the","diag","labels","kj","jk")

corrplot(M, type="upper",p.mat = p.mat, sig.level = 0.05)

    Output I got:Output