display p-values and significance levels
herwig wrote:
Hi there,
I got a piece of code for the Iris data which allows to display correlation
coefficients for each Iris species in the lower panel (color coded). I would
now like to add e.g. a "*" to show the significance of each correlation next
to the correlation coefficient.
Furthermore I would like to make a t.test between the species "setosa" and
"versicolor" for Sepal.length, ..width..... and so on and display it int
the horizontal panel
I would appreciate any help,
Herwig
here the code I've got at the moment:
panel.cor <- function(x, y, digits=2, prefix="", splitvar, col.cor, ...)
{
usr <- par("usr"); on.exit(par(usr))
par(usr = c(0, 1, 0, 1))
r <- abs(cor(x, y))
if(!missing(splitvar)) {
r <- c(r, abs(sapply(lapply(split(cbind.data.frame(x, y),
splitvar), cor), function(x)x[1,2])))
}
txt <- format(c(r, 0.123456789), digits=digits)[1:4]
txt <- paste(prefix, txt, sep="")
if(missing(col.cor)) col.cor <- c("black", "red", "green3", "blue")
for(i in 1:length(txt)) {
text(0.5, (1/(length(txt)+1))*i, txt[i], col = col.cor[i])
}
}
pairs(iris[1:4], main = "Anderson's Iris Data -- 3 species",
pch = 21, bg = c("red", "green3", "blue")[unclass(iris$Species)],
lower.panel=panel.cor, splitvar=iris$Species)
------------
#here the code I tried to implement to show the significance levels
test <- cor.test(x, y, ,use="complete.obs")
Signif <- symnum(test$p.value, corr = FALSE, na = FALSE,
cutpoints = c(0, 0.001, 0.01, 0.05, 0.1, 1),
symbols = c("***", "**", "*", ".", " "))
Not sure if this is homework...
first part of your question is implemented as follows (including some
other code improvements), just use the same ideas for the t-tests ....
panel.cor <- function(x, y, digits=2, prefix="", splitvar, col.cor, ...)
{
usr <- par(usr = c(0, 1, 0, 1))
on.exit(par(usr))
r <- abs(cor(x, y))
test <- cor.test(x, y, use="complete.obs")[["p.value"]]
if(!missing(splitvar)){
temp <- split(data.frame(x=x, y=y), splitvar)
r <- c(r, abs(sapply(temp, function(x) cor(x)[1,2])))
test <- c(test, sapply(temp, function(x) cor.test(x$x, x$y,
use="complete.obs")[["p.value"]]))
}
Signif <- symnum(test, corr = FALSE, na = FALSE,
cutpoints = c(0, 0.001, 0.01, 0.05, 0.1, 1),
symbols = c("***", "**", "*", ".", " "))
txt <- format(r, nsmall = digits, digits = digits)
txt <- paste(prefix, txt, " ", Signif, sep="")
if(missing(col.cor))
col.cor <- c("black", "red", "green3", "blue")
text(0.5, seq_along(txt) / (length(txt)+1), txt, col = col.cor,
adj=c(0, 0.5))
}
pairs(iris[1:4], main = "Anderson's Iris Data -- 3 species",
pch = 21, bg = c("red", "green3", "blue")[unclass(iris$Species)],
lower.panel=panel.cor, splitvar=iris$Species)
Uwe Ligges