Description Usage Arguments Details Value Author(s) See Also Examples
View source: R/test.diff.cor.R
This functions uses the Fisher Z-transform (atanh) to test the null hypothesis of no difference in correlations between x1 and y1 versus x2 and y2.
1 2 3 |
x1 |
numeric vector, x-values for the first sample. |
y1 |
numeric vector, y-values for the first sample. |
x2 |
numeric vector, x-values for the second sample. |
y2 |
numeric vector, y-values for the second sample. |
The alternative
argument specifies the alternative hypothesis given
below.
H0: cor(x1, y1) = cor(x2, y2) |
||
"two.sided" | => |
H1: cor(x1, y1) != cor(x2, y2) |
"greater" | => |
H1: cor(x1, y1) > cor(x2, y2) |
"less" | => |
H1: cor(x1, y1) < cor(x2, y2)
|
A numeric vector giving correlation for each group, size-estimate and standard error, confidence intervals and p-values.
Anders Ellern Bilgrau <anders.ellern.bilgrau (at) gmail.com>
Similar usage to cor.test
in stats
, however
not the same!
See test.diff.cor
for a vectorised version.
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 | x1 <- rnorm(100)
y1 <- rnorm(100)
x2 <- rnorm(110)
y2 <- 4*x2 + 0.5*rnorm(110) + 1
plot(x1, y1, xlim = range(x1, x2), ylim = range(y1, y2))
abline(lm(y1 ~ x1))
points(x2, y2, col = "red")
abline(lm(y2 ~ x2), col = "red")
diff.test <- correlateR:::test.diff.cor.single
round(data.frame(
two = diff.test(x1, y1, x2, y2, alternative = "two.sided"),
les = diff.test(x1, y1, x2, y2, alternative = "less"),
gre = diff.test(x1, y1, x2, y2, alternative = "greater")), 2)
round(diff.test(x1, y1, x1, y1, alternative = "two.sided"), 3)
round(diff.test(x1, y1, x1, y1, alternative = "less"), 3)
round(diff.test(x1, y1, x1, y1, alternative = "less"), 3)
|
Add the following code to your website.
For more information on customizing the embed code, read Embedding Snippets.