Information Technology Reference
In-Depth Information
AtMr
0.01
RAC
CmRb
0.1
0.02
0 .005
0.02
Gaut
0.05
0.01
0.01
CrJk
InAs
KZN
DrgR
DrgR
CmRb
AtMr
0.004
0.2
InAs
Mrd
CmAs
0
0
0.002
0.1
PubV
WCpe
Arsn
CmAs
BRs
Rape
0.02
BNRs
Mpml
NW s t
FrSt
0.05
0.002
ECpe
0.01
Limp
0.1
NCpe
0.004
0.01
0.001
0.02
0.15
AGBH
PubV
Figure 7.19 Two-dimensional CA biplot of the 2007/08 crime data set. Approx-
imating the row profiles R 1 ( X - E ) by plotting R 1 / 2 U and C 1 / 2 V (case B)
with arguments ca.variant = RowProfB ; RowProf.scaled.markers = FALSE
(the default); lambda = TRUE . (Lambda evaluates to 672.66, indicating that setting
lambda = FALSE would result in a biplot in which all the row points are squeezed into
one another with the column points more spread out.) Calibrations on axes are as in
Figure 7.18.
Figure 7.27 is an example where a reflection and/or a rotation of the biplot might be
considered for comparison purposes. In Figure 7.27 we have made the following changes
in the arguments of cabipl: reflect = "y" , rotate = 70 . In the resulting biplot
shown in Figure 7.28 all the distance properties of the biplot have been retained but the
outcome is a biplot that is visually much easier to compare with previous biplots such
as Figure 7.5.
Search WWH ::




Custom Search