Base Graphics in R (part II)
STAT 133
Gaston Sanchez
CC BY-SA 4.0
Base Graphics
2 / 72
Graphics in R
Traditional Graphics
I
I
I
R "graphics" follows a static, ”painting on canvas”
model.
Graphics elements are drawn, and remain visible until
painted over.
For dynamic and/or interactive graphics, R is limited.
3 / 72
Traditional Graphics
Traditional graphics model
In the traditional model, we create a plot by first calling a
high-level function that creates a complete plot, and then we
call low-level functions to add more output if necessary
4 / 72
Dataset mtcars
head(mtcars)
##
##
##
##
##
##
##
Mazda RX4
Mazda RX4 Wag
Datsun 710
Hornet 4 Drive
Hornet Sportabout
Valiant
mpg cyl disp hp drat
wt qsec vs am gear carb
21.0
6 160 110 3.90 2.620 16.46 0 1
4
4
21.0
6 160 110 3.90 2.875 17.02 0 1
4
4
22.8
4 108 93 3.85 2.320 18.61 1 1
4
1
21.4
6 258 110 3.08 3.215 19.44 1 0
3
1
18.7
8 360 175 3.15 3.440 17.02 0 0
3
2
18.1
6 225 105 2.76 3.460 20.22 1 0
3
1
5 / 72
Scatter plot
# simple scatter-plot
plot(mtcars$mpg, mtcars$hp)
250
●
● ●
●
●
●
150
● ● ●
●●●
●●
● ●
●
●●
●
●
●
●
●
50
mtcars$hp
●
10
15
20
25
●
●
●
●
30
mtcars$mpg
6 / 72
Axis Labels
# xlab and ylab
plot(mtcars$mpg, mtcars$hp, xlab = "miles per gallon",
ylab = "horsepower")
250
●
● ●
●
●
●
150
● ● ●
●●●
●●
● ●
●
●
●●
● ●
●
●
50
horsepower
●
10
15
20
25
●
●
●
●
30
miles per gallon
7 / 72
Title and subtitle
# title and subtitle
plot(mtcars$mpg, mtcars$hp, xlab = "miles per gallon",
ylab = "horsepower", main = "Simple Scatterplot",
sub = 'data matcars')
Simple Scatterplot
250
●
● ●
●
●
●
150
● ● ●
●●●
●●
● ●
●
●
●●
● ●
●
●
50
horsepower
●
10
15
20
25
●
●
●
●
30
miles per gallon
data matcars
8 / 72
x and y coordinate ranges
250
●
●
●
●
● ●
●
150
● ● ● ●●●
●●
● ●
●●
●
● ●
●
●
●
50
mtcars$hp
350
# 'xlim' and 'ylim'
plot(mtcars$mpg, mtcars$hp, xlim = c(10, 35), ylim = c(50, 400))
10
15
20
25
●
●
30
●
●
35
mtcars$mpg
9 / 72
Type
250
150
50
mtcars$hp
# using 'type' (e.g. lines)
plot(mtcars$mpg, mtcars$hp, type = "l")
10
15
20
25
30
mtcars$mpg
10 / 72
Points
# character expansion 'cex'
# and 'point character'
plot(mtcars$mpg, mtcars$hp, cex = 1.5, pch = 1)
250
●
●
●●
●
●
150
● ●● ●●●
●●
● ●
●●
●
● ●
●
●
●
50
mtcars$hp
●
10
15
20
25
●
●
● ●
30
mtcars$mpg
11 / 72
Point symbols (pch) available in R
●
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
●
●
17
18
19
20
22
23
24
25
●
●
●
16
●
21
12 / 72
Point Character
# 'pch' can be any character
plot(mtcars$mpg, mtcars$hp, pch = "@")
250
@
@
@
@@
@
150
@ @@ @@@
@
@
@@
@
@ @
@@
@
@
@
50
mtcars$hp
@
10
15
20
25
@
@
@@
30
mtcars$mpg
13 / 72
Point Character
250
●
150
●
●
●
●
●
50
mtcars$hp
# 'pch' symbols will be recycled
plot(mtcars$mpg, mtcars$hp, pch = 1:25)
10
15
20
25
●
●
30
mtcars$mpg
14 / 72
Point Colors
# color argument 'col'
plot(mtcars$mpg, mtcars$hp, pch = 19, col = "blue", cex = 1.2)
250
●
●
●
●●
●
150
● ● ● ●●●
●●
● ●
●
●
●●
● ●
●
●
50
mtcars$hp
●
10
15
20
25
●
●
● ●
30
mtcars$mpg
15 / 72
Coloring Point Symbols
I
I
the col argument can be used to color symbols
symbols 21 through 25 can additionally have their
interiors filled by using the bg (background) argument
16 / 72
Coloring Point symbols
●
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
●
●
17
18
19
20
22
23
24
25
●
●
●
16
●
21
17 / 72
So far ...
# using plot()
plot(mtcars$mpg, mtcars$hp,
xlim = c(10, 35), ylim = c(50, 400),
xlab = "miles per gallon",
ylab = "horsepower",
main = "Simple Scatterplot",
sub = 'data matcars',
pch = 1:25, cex = 1.2, col = "blue")
18 / 72
100 150 200 250 300 350 400
●
●
●
●
●
●
50
horsepower
Simple Scatterplot
10
15
20
25
●
●
30
35
miles per gallon
data matcars
19 / 72
Low-Level Functions
20 / 72
High and Low level functions
I
I
I
Usually we call a high-level function
Most times we change the default arguments
Then we call low-level functions
21 / 72
Scatter plot
# simple scatter-plot
plot(mtcars$mpg, mtcars$hp)
# adding text
text(mtcars$mpg, mtcars$hp, labels = rownames(mtcars))
# dummy legend
legend("topright", legend = "a legend")
# graphic title
title("Miles Per Galon -vs- Horsepower")
22 / 72
Miles Per Galon −vs− Horsepower
● Bora
Maserati
300
a legend
200
150
●Merc
● ●
MercMerc
450SLC
450SE
450SL
● ●Firebird
● Dino
Hornet
Sportabout
Pontiac
Ferrari
●●
Dodge
AMC Challenger
Javelin
100
● 280
Merc●Merc
280C
●
Lotus Europa
● ●4
Mazda
RX4
Wag
Mazda
RX4
Drive
Volvo
142E
● Hornet
Valiant
●Corona
ToyotaDatsun
●
Merc
● 230
710
Porsche● 914−2
●128●Corolla
Toyota
Fiat ●X1−9
Fiat
Merc●240D
Honda● Civic
50
mtcars$hp
250
●
Ford Pantera
L
● Z28
● 360
Duster
Camaro
Chrysler●Imperial
●
Lincoln Continental
●
Cadillac Fleetwood
10
15
20
25
30
mtcars$mpg
23 / 72
Scatter plot
# simple scatter-plot
plot(mtcars$mpg, mtcars$hp, type = "n",
xlab = "miles per gallon", ylab = "horsepower")
# grid lines
abline(v = seq(from = 10, to = 30, by = 5), col = 'gray')
abline(h = seq(from = 50, to = 300, by = 50), col = ' gray')
# plot points
points(mtcars$mpg, mtcars$hp, pch = 19, col = "blue")
# plot text
text(mtcars$mpg, mtcars$hp, labels = rownames(mtcars),
pos = 4, col = "gray50")
# graphic title
title("Miles Per Galon -vs- Horsepower")
24 / 72
Miles Per Galon −vs− Horsepower
Maserati Bora
200
● Ford Pantera L
● Duster
Camaro
Z28
360
● Chrysler Imperial
Lincoln Continental
Cadillac Fleetwood
●
●
●
150
●
● Merc
●450SLC
Merc
Merc
450SE
450SL
● ●Hornet
●Pontiac
Ferrari
Sportabout
Firebird
Dino
AMC
Dodge
Javelin
Challenger
●●
100
●
● Merc
Merc
280C280
● Lotus Europa
Hornet142E
RX4
4 Drive
Wag
Volvo
Valiant ● ●Mazda
● Toyota
Corona
●
230
● Merc
Datsun
710
● Porsche 914−2
●
● Fiat X1−9
● Fiat
● 128
Toyota Corolla
Merc 240D
● Honda Civic
●
50
horsepower
250
300
●
10
15
20
25
30
miles per gallon
25 / 72
Test your knowledge
The function plot()
A) accepts any type of vector
B) is a generic function
C) works only for 1-D and 2-D objects
D) is designed to display scatterplots and boxplots
26 / 72
Low-level functions
Function
points()
lines()
abline()
segments()
arrows()
rect()
polygon()
text()
symbols()
legend()
Description
points
connected line segments
straight lines across a plot
disconnected line segments
arrows
rectangles
polygons
text
various symbols
legends
27 / 72
Drawing Points with points()
points(x, y, pch = int, col = str)
I
I
pch integer or string indicating type of point character
col color of points
28 / 72
# drawing points
plot(mtcars$mpg, mtcars$hp, type = "n")
points(mtcars$mpg, mtcars$hp)
250
●
●●
●
●
●
150
● ● ● ●●●
●●
● ●
●
●
●●
● ●
●
●
50
mtcars$hp
●
10
15
20
25
●
●
● ●
30
mtcars$mpg
29 / 72
Connected Line Segments
lines(x, y, lty = str, lwd = num, col = str)
I
I
lty specifies the line texture. It should be one of
"blank" (0), "solid" (1), "dashed"(2), "dotted" (3),
"dotdash" (4), "longdash" (5) or "twodash" (6).
lwd and col specify the line width and colour
30 / 72
250
150
50
mtcars$hp
# connected lines
plot(mtcars$mpg, mtcars$hp, type = "n")
lines(mtcars$mpg, mtcars$hp, type = "s", lwd = 2)
10
15
20
25
30
mtcars$mpg
31 / 72
Line Graph Options
The type argument can be used to produced other types of
lines
I type = "l" line graph
I type = "s" step function (horizontal first)
I type = "S" step function (vertical first)
I type = "h" high density (needle) plot
I type = "p" draw points
I type = "b" draw points and lines
I type = "o" over-plotting points and lines
32 / 72
Line Graph Options
type = 'l'
type = 's'
type = 'S'
type = 'h'
33 / 72
Connected Line Segments
x <- 2005:2015
y <- c(81, 83, 84.3, 85, 85.4, 86.5, 88.3,
88.6, 90.8, 91.1, 91.3)
plot(x, y, type = 'n', xlab = "Time", ylab = "Values")
lines(x, y, lwd = 2)
title(main = "Line Graph Example")
34 / 72
Connected Line Segments
86
84
82
Values
88
90
Line Graph Example
2006
2008
2010
2012
2014
Time
35 / 72
Drawing Straight Lines
abline(a = intercept, b = slope)
abline(h = numbers)
abline(v = numbers)
I
I
I
The a / b form specifies a line in intercept / slope form
h specifies horizontal lines at given y-values
v specifies vartical lines at given x-values
36 / 72
# drawing straight lines
plot(mtcars$mpg, mtcars$hp, type = "n")
abline(v = seq(10, 30, by = 5), h = seq(50, 300, by = 50))
points(mtcars$mpg, mtcars$hp, pch = 19, col = "red")
250
●
●●
●
●
●
150
● ● ● ●●●
●●
● ●
●
●
●●
● ●
●
●
50
mtcars$hp
●
10
15
20
25
●
●
● ●
30
mtcars$mpg
37 / 72
Drawing Disconnected Lines
Disconnected lines can be drawn with the function:
segments(x0, y0, x1, y1)
I
I
The x0, y0, x1, y1 arguments give the start and end
coordinates of the segments.
Line texture, colour and width arguments can also be
given.
38 / 72
Drawing Line Segments
n <- 11
theta <- seq(0, 2 * pi, length = n + 1)[1:n]
x <- sin(theta)
y <- cos(theta)
v1 <- rep(1:n, n)
v2 <- rep(1:n, rep(n, n))
plot(x, y, type = 'n')
segments(x[v1], y[v1], x[v2], y[v2])
39 / 72
1.0
0.5
y
0.0
−0.5
−1.0
−1.0
−0.5
0.0
0.5
1.0
x
40 / 72
Drawing Polygons
Polygons can be drawn with the function:
polygon(x, y, col = str, border = str)
I
I
I
I
x, y give the coordinates of the polygon vertexes. NA
values separate polygons.
col specifies the color of the interior.
border specifies the color of the border.
line texture and width specifications can also be given
41 / 72
Drawing Polygons
0.04
0.02
0.00
Density
0.06
Kernel Density Curve
10
20
30
40
N = 32 Bandwidth = 2.477
42 / 72
Adding Text
We can add text using the function:
text(x, y, labels, ...)
I
I
I
I
I
I
x, y give the coordinates of the text.
labels gives the actual text strings.
font optional font for the text.
col optional color for the text.
srt rotation of the text.
adj justification of the text.
43 / 72
Drawing Text
plot(0.5, 0.5, xlim = c(0, 1), ylim = c(0, 1), type = 'n')
abline(h = c(.2, .5, .8),
v = c(.5, .2, .8), col = "lightgrey")
text(0.5, 0.5, "srt = 45, adj = c(.5, .5)",
srt = 45, adj = c(.5, .5))
text(0.5, 0.8, "adj = c(0, .5)", adj = c(0, .5))
text(0.5, 0.2, "adj = c(1, .5)", adj = c(1, .5))
text(0.2, 0.5, "adj = c(1, 1)", adj = c(1, 1))
text(0.8, 0.5, "adj = c(0, 0)", adj = c(0, 0))
44 / 72
0.8
1.0
Drawing Text
adj = c(0, 0)
j=
0.5
c(
.5
,
0.6
.5
)
adj = c(0, .5)
t=
adj = c(1, .5)
0.0
sr
0.4
0.2
45
,
ad
adj = c(1, 1)
0.0
0.2
0.4
0.6
0.8
1.0
0.5
45 / 72
Adding a Legend
Legends can be added with:
legend(xloc,
lty =
pch =
xjust
I
I
yloc, legend = text
linetypes, lwd = linewidths,
glyphname, col = colours,
= justification, yjust = justification)
xloc and yloc give the coordinates where the legend is
to be placed
xjust and yjust give the justification of the legend box
with respect to the location.
46 / 72
Adding Legends
# coords of exact line
plot(mtcars$mpg, mtcars$hp)
legend("topright", legend = "A legend")
●
300
A legend
200
●
●
150
●
● ●
●●●
●●
100
●
●
●●
●
●
●
●
●
●
50
mtcars$hp
250
●
● ●
●
●
●
●
●
10
15
20
25
30
mtcars$mpg
47 / 72
Plots from scratch
48 / 72
Customizing Annotations
It is also possible to create a plot from scratch. Although this
procedure is less documented, it is extremely flexible:
1. call plot.new() to start a new plot frame
2. call plot.window() to define coordinates
3. then call low-level functions:
4. typical options involve axis()
5. then title() (title, subtitle)
6. after that call other function: e.g. points(), lines(),
etc
49 / 72
Plot from Scratch
plot.new()
plot.window(xlim = c(0, 10), ylim = c(-2, 4), xaxs = "i")
axis(1, col.axis = "grey30")
axis(2, col.axis = "grey30", las = 1)
title(main = "Main Title",
col.main = "tomato",
sub = "Plot Subtitle",
col.sub = "orange",
xlab = "x-axis", ylab = "y-axis",
col.lab = "blue", font.lab = 3)
box("figure", col = "grey90")
50 / 72
Plot from Scratch
Main Title
4
3
y−axis
2
1
0
−1
−2
0
2
4
6
8
10
x−axis
Plot Subtitle
51 / 72
Another plot from scratch
set.seed(5)
x <- rnorm(200)
y <- x + rnorm(200)
plot.new()
plot.window(xlim = c(-4.5, 4.5), xaxs = "i",
ylim = c(-4.5, 4.5), yaxs = "i")
z <- lm(y ~ x)
abline(h = -4:4, v = -4:4, col = "lightgrey")
abline(a = coef(z)[1], b = coef(z)[2], lwd = 2, col = "red")
points(x, y)
axis(1)
axis(2, las = 1)
box()
title(main = "A Fitted Regression Line")
52 / 72
A Fitted Regression Line
●●
4
●
●
●
●
●
●
●
● ●●
●
●
● ●
●
●
●
● ● ●
● ● ●
● ●
● ●●●
● ● ● ●●●● ●●
●
●●●● ● ●
● ●
● ● ●●
● ●●
●
● ●●●
●
● ●●●●●●
●
● ● ●
●●
●
●●
●
● ●●
●
●● ● ●
●● ●
●
●● ●● ●
●● ● ●
● ●● ●
● ● ●●
● ●●
●
●
●
● ●●●● ●
● ●● ●●●
● ●●●●
●
● ●
●●
●
●
●● ● ●
● ● ● ●
● ●●
●●●●
●
●
●●●
●
●
● ● ●
●
●
●
●●
●●
● ● ● ●
● ●
●
●
●
●
●
●
●● ●
●
●
2
0
−2
●
●
●
●
●
●
−4
●
−4
−2
0
2
4
53 / 72
Creating a Plot from Scratch
I
Start a new plot with plot.new()
I
plot.new() opens a new (empty) plot frame
I
plot.new() chooses a default plotting region
54 / 72
Setting Up Coordinates
Then use plot.window() to set up the coordinate system for
the plotting frame
# axis limits (0,1)x(0,1)
plot.window(xlim = c(0, 1), ylim = c(0, 1))
By default plot.window() produces axis limits which are
expanded by 6% over those actually specified.
55 / 72
Setting Up Coordinates
The default limits expansion can be turned-off by specifying
xaxs = "i" and/or yaxs = "i"
plot.window(xlim, ylim, xaxs = "i")
56 / 72
Aspect Ratio Control
Another important argument is asp, which allows us to specify
the aspect ratio
plot.window(xlim, ylim, xaxs = "i", asp = 1)
asp = 1 means that unit steps in the x and y directions
produce equal distances in the x and y directions on the plot.
(Important to avoid distortion of circles that look like ellipses)
57 / 72
Drawing Axes
The axis() function can be used to draw axes at any of the
four sides of a plot.
I side=1 below the graph
I side=2 to the left of the graph
I side=3 above the graph
I side=4 to the right of the graph
58 / 72
Customizing Axes
Axes can be customized via several arguments (see ?axis)
I location of tick-marks
I labels of axis
I colors
I sizes
I text fonts
I text orientation
59 / 72
Plot Annotation
The function title() allows us to include labels in the
margins
I main main title above the graph
I sub subtitle below the graph
I xlab label for the x-axis
I ylab label for the y-axis
60 / 72
Customizing Annotations
The annotations can be customized with additional arguments
for the fonts, colors, and size (expansion)
I font.main, col.main, cex.main
I font.sub, col.sub, cex.sub
I font.lab, col.lab, cex.lab
61 / 72
Drawing Arrows
Arrows can be drawn with the function:
arrows(x0, y0, x1, y1, code = int,
length = num, angle = num)
I
I
I
The x0, y0, x1, y1 arguments give the start and end
coordinates.
code=1 head at the start, code=2 head at the end,
code=3 head at both ends
length of the arrow head and angle to the shaft
62 / 72
Drawing Arrows
plot.new()
plot.window(xlim = c(0, 1), ylim = c(0, 1))
arrows(.05, .075, .45, .9, code = 1)
arrows(.55, .9, .95, .075, code = 2)
arrows(.1, 0, .9, 0, code = 3)
text(.5, 1, "A", cex = 1.5)
text(0, 0, "B", cex = 1.5)
text(1, 0, "C", cex = 1.5)
63 / 72
Drawing Arrows
A
B
C
64 / 72
Drawing Rectangles
Rectangles can be drawn with the function:
rect(x0, y0, x1, y1, col = str, border = str)
I
I
I
x0, y0, x1, y1 give the coordinates of diagonally
opposite corners of the rectangles.
col specifies the color of the interior.
border specifies the color of the border.
65 / 72
Drawing Rectangles
# barplot manually constructed
plot.new()
plot.window(xlim = c(0, 5), ylim = c(0, 10))
rect(0:4, 0, 1:5, c(7, 8, 4, 3),
col = "turquoise",
border = "white")
axis(1)
axis(2, las = 1)
66 / 72
Drawing Rectangles
10
8
6
4
2
0
0
1
2
3
4
5
67 / 72
Plot Regions
68 / 72
Anatomy of Plot Frame and Region
Plot Region
margin 4
margin 2
margin 3
margin 1
69 / 72
Adjusting the Margins
Margins can be adjusted with the par() function in various
ways:
I In inches: par(mai = c(2, 2, 1, 1))
I In lines of text: par(mar = c(4, 4, 2, 2))
I Width and Height in inches: par(pin = c(5, 4))
70 / 72
One more scatter plot
# simple scatter-plot
op <- par(mar = c(5, 4, 3, 1))
plot(mtcars$mpg, mtcars$hp, type = "n",
xlab = "miles per gallon", ylab = "horsepower")
# grid lines
abline(v = seq(from = 10, to = 30, by = 5), col = 'gray')
abline(h = seq(from = 50, to = 300, by = 50), col = ' gray')
# points
points(mtcars$mpg, mtcars$hp, pch = 19, col = "blue")
# text (point labels)
text(mtcars$mpg, mtcars$hp, labels = rownames(mtcars),
pos = 4, col = "gray50")
# title
title("Miles Per Galon -vs- Horsepower")
# reset graphical margins
par(op)
71 / 72
Miles Per Galon −vs− Horsepower
Maserati Bora
200
● Ford Pantera L
● Duster
Camaro
Z28
360
● Chrysler Imperial
Lincoln Continental
Cadillac Fleetwood
●
●
●
150
●
● Merc
●450SLC
Merc
Merc
450SE
450SL
●●
●
Hornet
Pontiac
Ferrari
Sportabout
Firebird
Dino
AMC
Dodge
Javelin
Challenger
●●
100
●
● Merc
Merc
280C280
● Lotus Europa
Hornet142E
RX4
4 Drive
Wag
Volvo
Valiant ● ●Mazda
● Toyota
Corona
●
230
● Merc
Datsun
710
● Porsche 914−2
●
● Fiat X1−9
Merc 240D
●
50
horsepower
250
300
●
●
10
15
20
25
● Fiat
● 128
Toyota Corolla
Honda Civic
30
miles per gallon
72 / 72