Viewing Ecological data using R graphics

Transcription

Biostatistics
./
Illustrations in
Viewing Ecological data using R graphics
A.B. Dufour & N. Pettorelli
April 9, 2009
Presentation of the principal graphics dealing with discrete or continuous variables. Course proposed at the Institut of Zoology, London,
May 2009
Contents
1 Introduction
2
2 Principal graphics for one variable
2.1 Discrete variable . . . . . . . . . . . . . .
2.1.1 Already summarized information
2.1.2 Global data . . . . . . . . . . . . .
2.2 Continuous variable . . . . . . . . . . . .
.
.
.
.
2
2
2
3
4
3 Relationships between two variables
3.1 Two discrete variables . . . . . . . . . . . . . . . . . . . . . . . .
3.2 Two continuous variables . . . . . . . . . . . . . . . . . . . . . .
3.3 One discrete and one continuous variables . . . . . . . . . . . . .
7
7
11
14
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
4 Adding an information: two continuous variables and a factor 15
5 Conclusion
17
6 Appendice: Types of point and line
18
6.1 Point Types . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18
6.2 Line Types . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18
1
1
Introduction
Data viewing is an important first step to data modelling - think for example about detecting outliers and the possible influence of these values on your
analysis.
Many R packages contain graphic options but we will here focus on two libraries.
The first one MASS is included as a bunddle in the bases; the second one (ade4)
allows us introducing graphics which are specific to multivariate analysis.
library(MASS)
library(ade4)
2
2.1
Principal graphics for one variable
Discrete variable
We will here differentiate two situations, according to the type of data - either
you are trying to represent data which have been already summarized (e.g., into
vector (e.g.,
frequencies); or the data you are considering are organised in a
vector of binomial values).
2.1.1
Already summarized information
This represents the most common situation.
Example. The dataset ’Titanic’ contains the information on the fate of the passengers. We are interested in representing the number of passengers belonging
to the different classes who survived the incident.
library(MASS)
data(Titanic)
titanclas <- apply(Titanic, 1, sum)
Based on this information, we can compute two main graphics: the pie chart
(pie) and the bar plot (barplot).
par(mfrow = c(1, 2))
pie(titanclas)
pie(titanclas, main = "Passengers of Titanic", col = rainbow(4))
Passengers of Titanic
2nd
2nd
1st
3rd
Crew
1st
3rd
Crew
Logiciel R version 2.8.1 (2008-12-22) – course1.rnw – Page 2/19 – Compilé le 2009-04-09
Maintenance : S. Penel, URL : http://pbil.univ-lyon1.fr/R/pdf/course1.pdf
barplot(titanclas)
barplot(titanclas, main = "Passengers of Titanic", col = rainbow(4))
800
600
400
200
0
0
200
400
600
800
1st
2nd
3rd
Crew
1st
2nd
3rd
Crew
The most adapted graph is the barplot below since (1) the levels are ordered
and (2) the colors follow the class gradient.
barplot(titanclas, main = "Passengers of Titanic", col = c("white",
"yellow", "orange", "darkred"))
0
200
400
600
800
1st
2nd
3rd
Crew
Exercise. Create a graph to represent, using the same data set, the age distribution of the survivors.
2.1.2
Global data
The data are presented as a vector (one column, n individuals). Each element
of the vector contains the modality for the considered individual.
Example. The dataset crabs contains information on morphological measurements of 200 crabs belonging to two species (orange and blue). Information has
been collected on both sex.
data(crabs)
varspecies <- summary(crabs$sp)
pie(varspecies, col = c("blue", "orange"))
B
O
Exercise. Create a graph to represent, using the same data set, the distribution
of the measurements according to gender.
2.2
Continuous variable
We will here focus on:
1. the histogram hist
2. the boxplot boxplot
3. the Cleveland dot plot dotchart
4. a minor one: the one dimension scatter plot stripchart
Example. The aim of the following example is to represent the distribution of
the carapace length (CL) of the orange crabs.
caleor <- crabs$CL[crabs$sp == "O"]
hist(caleor, main = "Histogram")
boxplot(caleor, main = "Boxplot")
dotchart(caleor, main = "Cleveland dot plot")
stripchart(caleor, main = "Stripchart")
20
5 10
30
20
40
Boxplot
0
Frequency
Histogram
15
25
35
45
caleor
Cleveland dot plot
●
●●●
●
●●● ●
●
●●●
●
●
●
●
●●
●
●●
●●
●● ●
● ●●
●
●
●
● ●●
● ●●
●●●●
●
●
●●
●●
●
●●
●● ●
20
25
30
● ●
●●● ●
●
●●
●
●
●
●
●●
● ●●
●
●
●● ●
●● ●
●
●
●
●
●●
●
●●
35
Stripchart
40
45
●
●
20
25
30
35
40
45
The histogram can be improved using probability distribution (i.e. the superimposition of a theoretical distribution). We can also modified the breaks (see
hist.default for all the histogram information).
hist(caleor, freq = FALSE, main = "")
brex <- seq(10, 50, by = 10)
brex
[1] 10 20 30 40 50
hist(caleor, breaks = brex, freq = FALSE, main = "", col = grey(0.8))
0.00
0.02
Density
0.02
0.00
Density
0.04
0.04
15
25
35
45
10
caleor
20
30
40
50
caleor
The boxplot is based on the quartiles of the distribution. The function rug
allows to see each point of the distribution.
boxplot(caleor, col = grey(0.8), main = "carapace lengths", las = 1)
rug(caleor, side = 2)
carapace lengths
45
40
35
30
25
20
It’s also possible to add the median confidence interval on the graph (see boxplot.stats for more information).
boxplot(caleor, notch = TRUE, col = grey(0.8), main = "carapace lengths",
las = 1)
rug(caleor, side = 2)
carapace lengths
45
40
35
30
25
20
3
Relationships between two variables
3.1
Two discrete variables
The graphics used to represent the relationships between two discrete variables
are:
? mosaicplot()
? table.cont()
Example 1. In Caithness (Scotland), hair and eye colours were determined for
5387 inhabitants.
data(caith)
caith
fair red medium dark black
blue
326 38
241 110
3
light
688 116
584 188
4
medium 343 84
909 412
26
dark
98 48
403 681
85
mosaicplot(caith, main = "Colours of Hair and Eye for 5387 inhabitants of Caithness")
Colours of Hair and Eye for 5387 inhabitants of Caithness
light
medium
dark
black
dark
medium
red
fair
blue
mosaicplot(caith, main = "Colours of Hair and Eye for 5387 inhabitants of Caithness",
shade = T)
Colours of Hair and Eye for 5387 inhabitants of Caithness
light
medium
dark
<−4
Standardized
Residuals:
black
dark
medium
−4:−2
red
−2:0
0:2
fair
2:4
>4
blue
table.cont(caith, csize = 2)
dark
medium
light
blue
black
dark
medium
red
fair
Some variables can however have a lot of modalities and the graphics don’t help
to understand the data structure. In such cases, multivariate data analysis, such
as correspondence analysis, is the best way to view the data.
Example 2. The numbers of waders were counted for 19 species at 15 sites in
South Africa.
data(waders)
waders
S1
S2
S3 S4
S5
S6
S7 S8 S9 S10 S11 S12 S13
S14 S15
S16 S17
A
12 2027
0
0 2070
39 219 153
0 15 51 8336 2031 14941
19 3566
0
B
99 2112
9 87 3481 470 2063 28 17 145 31 1515 1917 17321 3378 20164 177
C 197 160
0
4 126
17
1 32
0
2
9 477
1
548
13
273
0
D
0
17
0
3
50
6
4
7
0
1
2
16
0
0
3
69
1
E
77 1948
0 19 310
1
1 64
0 22 81 2792 221 7422
10 4519
12
F
19 203
48 45
20 433
0
0 11 167 12
1
0
26 1790 2916 473
G 1023 2655
0 18 320
49
8 121
9 82 48 3411
14 9101
43 3230 587
H
87 745 1447 125 4330 789 228 529 289 904 34 1710 7869 2247 4558 40880 7166
I 788 2174
0 19 224 178
1 423
0 195 162 2161
25 1784
3 1254
0
J
82 350 760 197 858 962
10 511 251 987 191
34
87
417 4496 15835 5327
K 474 930
0 10 316 161
0 90
0 39 48 1183 166 4626
65
127
4
L
77 249 160 136 999 645
15 851 101 723 266 495
83 1253 1864 4107 1939
M
22 144
0
4
1
1
0 10
0
2
9 125
5
411
0
3
0
N
0 791
0
0
4
38
1 56
1 30 54
95
0 1726
0
0
0
O
0 360 128 43 364 1628
63 287 328 641 850
83
67
48 6499 9094 5647
S18 S19
A
5
0
B 1759
53
C
0
0
D
0
0
E
0
0
F 658
55
G
10
5
H 1632 498
I
0
0
J 1312 1020
K
0
0
L 623 527
M
0
0
N
0
0
O 1333 582
Understanding the structure of such a data set by just looking at the contingency
table is clearly impossible. If we try to graphically represent such data set, we
obtain:
mosaicplot(waders, main = "Wader Species", shade = T)
Wader Species
B
CD E F G
H
I
J
K L MN
O
S19
S18
S17 S16 S15
Standardized
Residuals:
S14
<−4
−4:−2
S13
−2:0
S12
0:2
2:4
>4
S11
S10
S9S8S7S6 S5 S4
S3 S2 S1
A
Alternatively, one can try table.cont.
table.cont(waders)
O
N
M
L
K
J
I
H
G
F
E
D
C
B
A
S19
S18
S17
S16
S15
S14
S13
S12
S11
S10
S9
S8
S7
S6
S5
S4
S3
S2
S1
3.2
Two continuous variables
A typical way of representing the relationship between two continuous variables
is the scatter plot.
Example 1. Let’s study, for instance, the relationship between the length (in
mm) and the width (in mm) of the carapaces, focusing here on the orange crab
species.
cawior <- crabs$CW[crabs$sp == "O"]
plot(caleor, cawior)
plot(cawior ~ caleor, main = "Length and Width of carapaces")
plot(cawior ~ caleor, pch = 20)
Length and Width of carapaces
●
● ●
●
50
●
●
●●
●●
●
45
● ●●
● ●● ●
40
●
●
●
●●
●
●●
●
● ●●
●
●
●●
●
●
●●
●
●● ●
●●
●
● ●
●
●●●
●
●● ●
35
●
●
●
●●●
●
●●
●●
●
●
●●
●
●
30
●
●
●
●●
●
● ●●
●
●●
●
●
●
●
●
●
●
25
50
45
40
●
●●
●
●
20
cawior
35
30
20
25
50
45
40
35
cawior
30
25
20
●
●●
●
●
●●
●●
●
●●
●
●●
●●
●
●●
●
●
●
●
●
●●
●
●●
●
●●
●
●
●●
●●●
●●
●
●
●●
●
●●
●
●
●● ●
●
●
●●
●
●
●
●●
●
●●
●
●
●
●
●
●
●
●
●
●
● ●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
cawior
●
●●
●
●
●●
●●
●
●●
●
●●
●●
●
●●
●
●
●
●
●
●●
●
●●
●
●●
●
●
●●
●●●
●●
●
●
●●
●
●●
●
●
●● ●
●
●
●●
●
●
●
●●
●
●●
●
●
●
●
●
●
●
●
●
●
● ●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
20 25 30 35 40 45
20 25 30 35 40 45
20 25 30 35 40 45
caleor
caleor
caleor
Notice that if we want to add some information on the graph, we first need to
compute a plot().
plot(cawior ~ caleor, main = "Length and Width of the 100 carapaces",
pch = 20, xlab = "length (in mm)", ylab = "width (in mm)")
abline(h = mean(cawior), lty = 2, col = grey(0.6))
abline(v = mean(caleor), lty = 2, col = grey(0.6))
points(mean(caleor), mean(cawior), col = "red", pch = 20, cex = 2)
Length and Width of the 100 carapaces
●
●
●
50
●
●
●●
40
●
●
●
●
●● ● ●
●●
●
● ● ●
●●
●●
●
●
● ●●
35
●
● ●
●●
●●
●
●
●●
●
●●
●●
●
● ●
●●
●
●
● ●
●●●
● ●
●●●
●
●●
●
30
width (in mm)
45
● ●
●
●
● ●●
●
●
●
● ●
●
●●
● ●
●
●
●●
●
●
●
●
25
●
●
●●
●
20
●
●
20
25
30
35
40
45
length (in mm)
Exercise. Choose two other continuous variables in this data set and draw the
best graph.
Sometimes, however, several individuals can get the same values (especially in
the case of fixed values). In the classical plot, these individuals will only be
represented by one single point. So if we want to see the superimposition of
points, we need to use specific arguments in the previous graph or built a new
one such as the sunflowerplot.
Example 2. Let’s examine here the famous data set used by W.S. Gosset (aka
Student) to develop the t-test: the body heights and the mid-finger lengths of
3000 criminals.
data(crimtab)
table.cont(crimtab)
13.5
13.4
13.3
13.2
13.1
13
12.9
12.8
12.7
12.6
12.5
12.4
12.3
12.2
12.1
12
11.9
11.8
11.7
11.6
11.5
11.4
11.3
11.2
11.1
11
10.9
10.8
10.7
10.6
10.5
10.4
10.3
10.2
10.1
10
9.9
9.8
9.7
9.6
9.5
9.4
crimtab.dft <- as.data.frame(crimtab)
expand.dft <- function(x, na.strings = "NA", as.is = FALSE, dec = ".") {
DF <- sapply(1:nrow(x), function(i) x[rep(i, each = x$Freq[i]),
], simplify = FALSE)
DF <- subset(do.call("rbind", DF), select = -Freq)
for (i in 1:ncol(DF)) {
DF[[i]] <- type.convert(as.character(DF[[i]]), na.strings = na.strings,
as.is = as.is, dec = dec)
}
DF
}
crimtab.raw <- expand.dft(crimtab.dft)
x <- crimtab.raw[, 1]
y <- crimtab.raw[, 2]
plot(x, y, las = 1, main = "3000 criminals", ylab = "Body height [cm]",
xlab = "Left mid finger [cm]")
3000 criminals
●
190
●
●
●
●
●●
●
●●●●
●
●
●●●●●●●●●●●●●●
180
Body height [cm]
●
●
●●●●●
●●●●●●●●●●●●●●
●●
●
●
●●●●●●●●●●●●●●●●●●●
●
●●●●●●●●●●●●●●●●●●●●●●●●●
●
170
●●●●●●●●●●●●●●●●●●●●●●●
●●●●●●●●●●●●●●●●●●●●●
●
●
●●●●●●●●●●●●●●●●●●
●●●●●●●●●●●●●●●●●
●●●●●●●●●●●●●●
150
●●●●●●●●●
●
●
●●●●●●●●●●●●●●●●●●●●●
●●●●●●●●●●●●●●●●●●●●●●
●●●
●
●
●●●●●●●●●●●●●●●●●●●●●●●●
●
160
●●
●●
●●●
●
●
●
●●
●
●
●●
●●●●
●
●
●
10
11
12
13
Left mid finger [cm]
If we want to see all the points, one possible solution is to jitter the data.
plot(jitter(x), jitter(y), las = 1, main = "3000 criminals", ylab = "Body height [cm]",
3000 criminals
●
190
● ● ●
●● ● ●
●
●
●
●●
●●
●
● ●●
●● ●
●●
●
●
●
●
●
●
●●
●●●
● ● ●●
●
●●
●
●●
●●
●●
●
●
●●
●● ●
●
●
●
●
●
●●●●
●●
●●
●●
●●
●
●●
●
●
●
●
●●
●●
●
●
●
●
●
●
●
●
●
●
●●
●
●
●
●
●
●
●
●
●●
●
●
●
●
●
●
●●●
●
●
●
●
●
●
●
Body height [cm]
180
●
●
●
●
●●
●●
●●
●●
●
●
●
●
●
●●
●●
●
●
●
●●
●●
●
●
●
●
●
●
●
● ●●
●
●
●
●
●
●●
●
●●
●●
●
●●
●
●
●
●
●
●
●
●
●●
●●
●
●
●
●
●
●●
●
●
● ●●
●
●
●
●
●
●
●
●
●●
●
●
●
●
●
●
●
●
●
●●
●
●
●
●●
●
●●
●
●
●●
●●
●
●●
●
●●
●●
●
●
●
●
●
●
●
●
●●
●
●●
●
●
●●
●
●
●
●
●
●
●
●
●
●
●
●●
●●
●
●
●●
● ● ●●
●
●
●
●
●●
●●
●
●●
●
●
●
●
●●
●
●● ●
●
●
●
●●
●
●
●●
●
●
●●
●
●
●●
●
●●
●●
●
●
●
●
●
●
●
●
●
●
●
●●
●
●●
●
●
●●
●
●
●●
●
●
●
●
●
●
●
●
●
●
●●
●
●
●
●
●
●
●
●
●●
●
●
●●
●
●
●
●
●
●
●●
●
●
●
●●
●
●
●
●
● ● ● ●●
●
●
●
●
●
●
●
●
●
●
●
●
●●
●
●
●
●
●●
●
●
●
●
● ● ●●
●
●
●
●
●●
●
●
●●
●
●
●
●
●
●
●●
●●
●
●●
●
●
●
●
●
●
●
●
●
●
●
●
●●
●
●
●●
●●
●
●
●
●
●
●●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●●
●
●
●
●●
●
●
●
●●
●
●
●
●
●
●
●
●
●
●●
●
●
●●
●
● ●
●
●●
●●
●
●
●
●
●● ●
●
●
●●
●
●
●
●
● ●●
●●
●
●
●
●●
●
●
●
●
●
●
●
●
●
●●
●
● ●● ●
●
●
●
●
●
●●
●
●
●
●
●●
●
●
●
●●
●
●
●
●
●
●
●
●
●
●
●
●●
●
●
●
●
●●
●
●
●
●
●
●●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●●●
●
●
●●
●
●
●
●
●
●
●
●●
●
●
●
●
●
●
●
●
●●
●●
●●
●● ●
●
●
●
●
●
●
●●
●
●
●●
●
●
●
●
●
●
●●
●
●●
●
●●
●●
●
●
●
●
●
●●
●
●
●
●
●●
●
●
●●
●
●
●●
●
●
●●
●
●● ● ●
●
●
●●
●
●
●
●
●
●
●
●
●
●●
●
●
●
●
●
●
●
●●
●
●
●
●
●
●●
170
●
●
160
●
150
●
●
●
●
●
●
●●
●
●
●
●
●
●
●●
●●
●●
●
●
●
●
●
● ●● ●
●
●
●
●
●
●●
●
●
●●
●
●
●
●
●●
●
●
●
●
●●
●
●
●
●
●
●●
●●
●
●●
●●
●
●
●
●
●●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●●
●●
●
●
●
●
●●
●●●
●
●
●
●●
●
●
●
●●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
● ●●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●●
●●
●●
●
●●
●
●
●
●
●●
●
●
●
●
●
●
●
●
●
●●
●●
●
●
●●
●
●
●
●
●
●
●
● ●●● ●●
●
●
●
●
●
●
●
●●
●●
●●
●
●●
●●
●●
●●
●
●
●
●
●
●
●●
●
●
●
●
●
●●
●
●
●
●
●
●
●
●
●●
●
●●
●
●
●●
●
●
●●
●
●
●
●●
●
●
●●
●
●●
● ●●
●
●●
●
●
●●
●
●●
● ●
●
●
●
●
●
●●
●
●
●
●●
●
●
●
●
●
●●
●
●●
●●
●
●
●
●●
●
●●
●
●
●
●
●●
●
●
●
●●
●
●
●●
●
●
●●
●
●
●
●
●
●●
●
●
●
● ●
●
●●
●
●●
●
●●
●
●
●
●
●●
●●
● ●●
●
●
●
●●
●
●●
●
● ●
●●
●
●●
●●
●
●●
●● ●
●
●
●
●●
●●
●
●
●●
●
●
●
●●
●
●●● ●
●●
●● ●
●
●●
● ● ●
●●
● ●
●●
●
●
●●● ●
● ●
●
●●
● ● ●●
●●
● ● ●●
● ● ● ●●
●●
●● ● ●
●
●● ● ●
●
●
●
140
10
11
12
13
Another option is the sunflowerplot. The number of repetitions is represented
by the number of petals around a point.
sunflowerplot(x, y, las = 1, main = "3000 criminals", ylab = "Body height [cm]",
3000 criminals
●
190
●
●
Body height [cm]
180
170
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
● ●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
● ● ●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
● ●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
● ●
●
●
● ● ●
●
●
●
●
●
●
● ●
●
●
●
●
●
●
●
● ●
●
●
● ●
●
●
●
●
●
●
● ●
●
●
●
●
●
●
●
●
●
●
160
150
●
●
●
●
●
●
● ●
●
●
● ●
●
●
●
● ●
●
●
●
●
●
● ●
● ●
●
●
●
● ●
●
●
●
●
10
11
12
13
Exercise. Is there any visible repetition when exploring the relationship between length and width of carapaces (focusing again on orange crabs)?
3.3
One discrete and one continuous variables
The main displays to link discrete and continuous variables are the boxplot or
the dotchart graphics.
Example. Let’s explore the relationship between carapace length and sex in
orange crabs.
45
40
35
30
25
20
20
25
30
35
40
45
crabsex <- crabs$sex[crabs$sp == "O"]
par(mar = c(3, 2, 2, 2))
boxplot(caleor ~ crabsex)
boxplot(caleor ~ crabsex, col = c("pink", "lightblue"), notch = TRUE)
dotchart(caleor, groups = crabsex)
dotchart(caleor, groups = crabsex, gdata = tapply(caleor, crabsex,
mean), pch = 20, gpch = 20, gcol = "red")
F
M
F
●
●
M
●
●
●
●●●
●
●●● ●
●
●●●
●
●●●
●
●
●
●
●
●●
●
●
●●●
●●●
F
●
●●●●
●
●●
●
●
●
●
●●
●
●
●
●●●
●
● ●
●● ●
●
●
●●
●
●
●
● ●
●●
●
●
●
●
●●
●
●
●
●
●●
●●
●
●
●
●●
●
●
●● ●
●● ●
20 25 30 35 40 45
M
F
●
●
●
M
●
●●
●
●
●●●
●●
●●
●●
● ●
●
●●
●
●
●●
●
● ●
●● ●
●
●● ●
●
●●●
● ●
●●
●
●
●
●●
●
●●
●
●
●
●●
●●
●●●
●
●●
● ●●
●●
●
●
●●
●●
●
●●
● ●
●
●
●●
●
●
● ●
●
●●
●
●
●
●
●
●● ●
●
●
20 25 30 35 40 45
Exercise. The data contains 4 other continuous variables. Produce a representation, in the same graphical device, of the relationship between each variable
and sex.
4
Adding an information: two continuous variables and a factor
We are here interested in computing three variables together (e.g., relationship
between sepal length and width of the famous Iris dataset, for the three species
- Iris setosa, Iris versicolor and Iris virginica). Two main representations can
be used: coplot and s.class of the ade4 package.
data(iris)
names(iris)
[1] "Sepal.Length" "Sepal.Width"
"Petal.Length" "Petal.Width"
"Species"
coplot(iris$Sepal.Width ~ iris$Petal.Length | iris$Species)
Given : iris$Species
virginica
versicolor
setosa
2
3
4
5
6
7
4.0
1
● ●
3.5
3.0
2.5
2.0
●
3.5
4.0
●
●
●
●
●
●
●●● ●
●
●
●●●●
●●●● ●
● ●
●●● ●
●●
● ●● ●
●
●
●
●
● ●●
● ●●
●● ●●● ●
●● ●●●
●● ●●●●
● ●●
●
●
● ●
●
●●
●
● ●●
●
● ●
●
●
●
2.5
3.0
iris$Sepal.Width
●
●●
● ●
● ● ● ●●
● ●●●
●● ●● ● ●● ●
●
●
●
●● ●
●
●
●
●●●
●
●
●
●
●
2.0
●
●
1
2
3
4
5
6
7
iris$Petal.Length
s.class(iris[, 1:2], iris$Species)
s.class(scale(iris[, 1:2], center = T, scale = F), iris$Species)
s.class(scale(iris[, 1:2], center = T, scale = F), iris$Species,
cstar = 0, cpoint = 0)
s.class(scale(iris[, 1:2], center = T, scale = F), iris$Species,
cellipse = 0)
d=2
d=1
●
●
●
●
●
●
●
● ●●
●●
●
●
●
●
● ●
●
● ●●
●
●
●
●● ●●● ●
● ●●
● ● ●
●●
●●●
● ●● ●●●
●●●● ●●
●●
●
●● ●●●●● ●
●
●●● ●●●●● ●
● ●
●
● ● ● ●●
● ●● ●
●
● ●
●●●
●
●
●
●
●
●
●
●
● ●
●
●
●
●
●
●●● ●
● ● ●●● ●
setosa
●●
● ●● ●
●
●
●
● ●
● ● ●
●
●
● ● ●
● ●
● ●
●
●
● ●
● ● ●
virginica
versicolor
setosa
●
● ●
●
●
●
●
●
●
● ●
●
●
● ●
● ● ●
●
●
●
●
● ●
● ● ●
● ● ● ●
● ●
● ●
● ● ● ● ●
●
●
● ● ●
● ● ● ● ●
●
●
●
●
●
● ●
●
● ●
●
● ● ●
●
●
●
●
●
●
●
●
●
virginica
versicolor
●
●
●
●
● ●
●
●
● ●
●
●
●
d=1
d=1
●
●
●
●
●
●
●
●
●
●
● ●
● ● ●
●
●
● ● ●
● ●
● ●
●
●
● ●
● ● ●
setosa
setosa
●
virginica
● ●
●
versicolor
●
●
●
●
●
●
●
● ●
●
●
●
●
● ●
●
●
● ●
● ● ●
●
●
●
●
● ●
● ● ●
● ● ● ●
● ●
● ●
● ● ● ● ●
●
●
● ● ●
● ● ● ● ●
●
●
●
●
●
● ●
●
● ●
●
● ● ●
●
●
●
●
●
●
●
●
●
virginica
versicolor
● ●
●
●
●
Exercise. The data set contains 3 species and 4 measurements. Produce a
representation of the relationship between two measurements and species.
5
Conclusion
Data viewing is a crucial first step to data modelling. Remembering functions
can be challenging.
provides some help such as:
1. help.start() opens a window with an interface for HTML help.
2. RSiteSearch("hist") looks for information on the
documentation, archives,. . . )
website (manuals,
3. help.search("hist") opens a page to inform where to find good explanations
4. help("hist") opens a page giving all the information about the research
object.
6
Appendice: Types of point and line
6.1
Point Types
●
●
13
2
8
14
20 ●
*
3
9
15
21
●
.
4
10
5
11
6
12
●
19
●
7
1
25
*
+
+
−
−
:
:
|
22
o
o
|
17
23
O
O
%
18
24
0
0
#
16
●
%
#
1. The point type is given by the pch argument. For instance, empty circles
(by default) is pch = 1 (see more information on the appendix of the
document).
2. The colour of a point is given by the col argument. For instance, we used
the red to compute the previous graphic using col = "red".
3. The inside colour of a point is given by the bg argument. For instance, we
used the yellow to compute the previous graphic using bg = "yellow".
4. For more details, look at the points() function.
6.2
Line Types
n <- 7
plot(0:n, 0:n, type = "n", xlim = c(0, n + 1), ylim = c(0, n + 1))
for (i in 1:n) {
lines(1:n, rep(i, n), lwd = 3, lty = i)
}
4
0
2
0:n
6
8
0
2
4
6
8
0:n
Line types are given by the lty argument. For instance, the continuous line type
(by default) is lty = 1 (see more information on the appendix of the document).
You may find useful to read the Graphical procedures chapter from An Introduction to R (following the menu Help, Manuals).

Viewing Ecological data using R graphics

Transcription

Documents pareils

Correspondence Analysis (COA or CA)

Press Kit 2015

insurance certificate

Living with the Regulation from a SME perspective

Exact Multivariate Tests of Asset Pricing Models with Stable

course5 ————— Linear Discriminant Analysis

titanium 2 (hard drive upgrade)

Stop-Motion Animation: Digital Storytelling in the Classroom

PM0502-13 594.46KB 2016-05

aircraft tracking - CCT

Licensing Information User Manual