Description Usage Arguments Details Value Author(s) Examples
Draws a correlation line, defined eigenvector of the first principle compoenent.
1 |
x |
The x variable |
y |
The y variable |
outlim |
Should the line be draw outside the limits of the data? |
... |
Other variables passed to |
This draws a 'correlation line' on a plot of data. A correlation line is distinct from a line drawn from a regression output (e.g. that produced by lm
) in that a correlation minimises residuals orthagonnaly to the line, not parallel with the y-axis (as a regression does). This is one of the main reasons why 'correlation doe snot equal causation'. There aren't a great number of good reasons to draw a correlation lin, but it is useful for teaching / demonstration.
If outlim
is FALSE (the default) only a line that fits inside the convex hull of x and y is drawn. If outlim
is true, a dshed line (lty=3
) is drawn first, and then the a line within the convex hull is drawn over this.
Draws a line on the exisiting plot object using a call to lines
Daniel Pritchard
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 | x <- c(-1.03, 0.79, -0.67, -0.1, 0.05, -0.76, 1.91, -0.52, -0.12,
0.07, 0.1, -1.92, 0.3, 1.42, -0.25, -0.23, -0.26, 1.19, 0.25,
-0.07, -1.55, 0.38, 1.86, 0.74, -0.48, 1.37, -0.73, 0.11, 0.5,
-2.32)
y <- c(0.59, 2.29, 0.36, -0.58, -0.86, 1.12, 0.09, -0.15, -1.47, 0.69,
-0.77, 1.23, -0.47, -0.17, -0.19, -0.84, -0.34, -0.47, 0.98,
-0.46, -0.84, 0.18, 0.59, -1.21, -0.83, 2.8, 0.71, 0.02, -1.18,
-0.82)
layout(matrix(c(1,2),nrow=1))
par(mai=c(0.8,0.8,0.4,0.4))
layout(matrix(c(1,2),nrow=1))
plot(x, y, main = "Regression", xlim=c(-2.5,2.5), ylim=c(-2,3), xaxs="i", yaxs="i")
yxlm <- lm(y ~ x)
lines(x, predict(yxlm), col = "red", lwd = 2)
segments(x, y, x, fitted(yxlm), col = "red")
plot(x, y, main = "Correlation", xlim=c(-2.5,2.5), ylim=c(-2,3), xaxs="i", yaxs="i")
corfit <- cor_line(x, y, outlim=TRUE, col='blue')
|
Add the following code to your website.
For more information on customizing the embed code, read Embedding Snippets.