Information Technology Reference
In-Depth Information
5
X
2
10
5
16
20
20
24
25
10
9
14
Z
15
25
6
20
3
15
7
15
10
17
8
4
1
18
19
15
11
22
23
13
21
10
12
20
5
0
25
Y
Figure 3.27
PCA biplot of the scaled data given in Table 3.18.
x and y. The plane of best fit, for PCA, will lie along the z-axis for the unscaled data,
while virtually any plane could be chosen for the scaled data.
3.7 Functions for constructing a PCA biplot
3.7.1 Function
PCAbipl
The main function for constructing PCA biplots, described in detail in this section, is
PCAbipl
. The reader is encouraged to study and experiment with the arguments of
PCAbipl
to appreciate all its capabilities. Several functions are called by
PCAbipl
for
drawing the biplot, adding features to it and changing its appearance; provision is made
for one-, two- and three-dimensional biplots. These functions are generally not directly
called by the user and will not be discussed in detail in this topic, but a complete set
of help files is available on the
Understanding Biplots
website. There are also several
functions closely related to
PCAbipl
that users may like to call to construct specific
types of PCA biplots or adding enhancements to an existing biplot. These functions are
briefly introduced in sections 3.7.2 - 3.7.9.
