Basic plotting with R#

Learning objectives

  • Questions:

    • How to plot in R

  • Objectives:

    • Learn basic plotting tools in R

    • Learn to export graphics to a file

  • Keypoints:

    • Plot

Data visualization:

  • Making exploratory graphs

    • To understand data properties

    • To find patterns in data

    • To suggest modeling strategies

    • To “debug” analyses

    • To communicate results

  • Principles of analytic graphics

  • Plotting systems and graphics devices in R

  • The base, lattice, and ggplot2 plotting systems in R

Plotting system

  • There are 3 main plotting systems in R:

    • The base plotting system

    • Start with blank plot and build up the plot

    • Plot is just a series of R command

    • Flexible

  • The Lattice plotting system (using package::lattice)

    • Plots created using single function call

    • Good for putting many plots to screen

    • Cannot add to plots once created

  • The ggplot plotting system (using package::ggplot2)

    • Similar to Lattice but easier

    • Many default mode

    • Flexible between Base and Lattice

image image image image image

  • R is a data analysis language, so naturally it comes with many built-in functions for plotting. Let’s look at some of them, in application to the mtcars data set.

data(mtcars)
dim(mtcars)
names(mtcars)
head(mtcars)
print(mtcars)

  • First, let’s make a bar plot of the miles-per-gallon values of the 32 cars. col specifies the color.

barplot(mtcars$mpg, col="green")

  • Now, let’s make a histogram for horsepower values.

hist(mtcars$hp, col="magenta")

  • Let’s display two jistorgrams on the same plot. We’ll use the function par to specify that the two histograms will be plotted stacked to each other, one on the left and one on the right.

par(mfrow=c(1,2))
hist(mtcars$mpg,col="blue")
hist(mtcars$wt,col="blue")

  • Let’s create a box plot for miles-per-gallon data. We’ll have to reset the plot options to let R know we don’t use multiple plots anymore; this is done with dev.off().

dev.off()
boxplot(mtcars$mpg,col="blue",main="Boxplot for mpg")

  • Now, let’s make this box plot for every value of cylinders (mtars$cyl). We will use the same function boxplot, but with a few more parameters. We will specify the data explicitly, as well as col (for colour scheme), main (main figure title), xlab and ylab (X- and Y-axis labels). We will use the function legend to create a legend. We call factors to find out the unique cylinder values.

factor(mtcars$cyl)
boxplot(mpg~cyl,data=mtcars,
        col=terrain.colors(3),main="MPG by car cylinders",
        xlab = "cylinders",ylab="mpg")
legend("topright",c("4","6","8"),fill = terrain.colors(3))

  • And a scatter plot. We’ll use the plot function for this purpose. The dots will be coloured according to the number of cylinders.

plot(mtcars$mpg, mtcars$wt, main="Car Fuel vs Weight",
     xlab="Mileage per Gallon",ylab="Weight",
     col = mtcars$cyl,pch=16,cex=3)
legend("topright",legend=c(8,6,4),pch=16,cex=3,
       col=c(8,6,4))

  • Here, pch is the plotting character; 16 corresponds to a circle, and cex is the size of the plotting characters. The full table of plotting characters is here:

img

  • What if we want to save a plot? Let’s save it as a PDF. Here’s how we do it:

pdf("myplot.pdf")
boxplot(mtcars$mpg,col="blue",main="Boxplot for mpg")
dev.off()

Graphics Devices

  • A graphics device is something where you can make a plot appear When you make a plot in R, it has to be “sent” to a specific graphics device.

  • A window on your computer (screen device): quick visualization

  • A PDF, PNG, JPEG file (file device) #recommended for documents, paper, presentation

  • The most common place for a plot to be “sent” is the screen device

    • On a Mac the screen device is launched with the quartz()

    • On Windows the screen device is launched with windows()

    • On Unix/Linux the screen device is launched with x11()

    • save graphic to PDF

    • save graphic to PNG, JPEG

dev.copy(png,"filename.png") # to save the image to file
dev.off() # to close all the graphical devices