TABLE OF CONTENTS The data that we discussed in Example 5.1 are also called dimensional measurement data because each variable is a measurement of length or distance. Every variable has the same practical or physical meaning as well as the same unit. In this section, we are going to discuss the special principal component analysis procedure developed by Yang (1996) to analyze this kind of data. Later we will show that this special procedure cannot only be used to analyze dimensional measurement data, but can also be used to analyze many multivariate data in which every variable has the same physical or practical meaning, for example, temperature measurement data at different surface locations of objects.
A well-known example of dimensional measurement data is the painted turtle example, which is discussed in almost every book that discusses principal component analysis.
Example 5.4 : The Painted Turtle Example Jolicoeur and Mosimann (1960) studied the relationship of size and shape for painted turtles. Table 5.4 contains the measurements on carapaces of 24 female turtles.
Table 5.4: Measurements of Turtle Dimensions Sample no. Length (mm)
x 1
Width (mm)
x 2
Height (mm)
x 3
1
98
81
38
2
103
84
38
3
103
86
42
4
105
86
42
5
109
88
44
6
123
92
50
7
123
95
46
8
133
99
51
9
133
102
51
10
133
102
51
11
134
100
48
12
136
102
49
13
138
98
