This is an early work-in-progress.
We can make simple scatterplots using the general plot
function. We use the vector of x values as the first parameter and the
vector of y values as the second parameter. Generally,
plot(explanatory, response)
Here are some examples.
(Of course, you'll generally get those values from an existing
data set.)
plot(c(1,2,4,5,5,7,10,11), c(2,3,4,3,5,6,7,8))
plot(c(1,2,2,3,3,3,4,4,5), c(1,5,10,2,6,11,3,7,12))
You can also join the two vectors together with the
~ operation, in which case the code will look
something like
plot(response ~ explanatory)
We'll tend to use whichever of these forms is most convenient at the
moment.
The car library provides somewhat more elegant scatterplots,
with the regression line and some box-and-whiskers plots added. To
use that library, you need to load it first.
library(car, lib="/home/rebelsky/Stats115/Packages")
If you're working on your own computer, you need to install your own copy
of that library.
install.package("car")
Once you've loaded that library, you can make scatterplots with
the scatterplot function. Typically, the first
parameter is a vector of X values and the second parameter is a
vector of Y values.
scatterplot(c(1,2,4,5,5,7,10,11), c(2,3,4,3,5,6,7,8))
scatterplot(c(1,2,2,3,3,3,4,4,5), c(1,5,10,2,6,11,3,7,12))
If you don't want to see the funky curve that attempts to show
a best approximation, add smooth=FALSE.
If you don't want to see the regression line, add
reg.line=FALSE.
