Information Technology Reference
In-Depth Information
Ta b l e 7 . 1
Listing of the function being approximated (model), the weights used in the weighted least-squares
criterion and the inner product approximation to the model used as a basis for plotting the variants of CA
discussed in this chapter. Matrices giving coordinates for typical biplots are given in the final column. The
variants are all special cases of the general result given in the final row of the table.
CA variant
Model
Weights
Inner product
Typical biplot
W
1
,
W
2
Approximation
R
−
1
/
2
C
−
1
/
2
V
1
/
2
,
V
1
/
2
Pearson
residuals
(
X
−
E
)
I
,
I
U
U
(case A)
U
,
V
(case B)
V
C
1
/
2
R
−
1
/
2
,
C
−
1
/
2
R
1
/
2
U
R
1
/
2
U
1
/
2
,
C
1
/
2
V
1
/
2
Independence
deviations
X
-
E
V
C
−
1
/
2
R
−
1
C
−
1
R
1
/
2
,
C
1
/
2
R
−
1
/
2
U
R
−
1
/
2
U
2
(case A)
R
−
1
/
2
U
,
C
−
1
/
2
V
(case B)
1
/
2
,
C
−
1
/
2
V
1
/
Contingency
ratio
(
X
−
E
)
R
−
1
(
X
−
E
)
C
−
1
/
2
R
−
1
/
2
R
1
/
2
,
I
I
,
C
1
/
2
R
−
1
/
2
U
V
U
V
C
−
1
/
2
R
−
1
/
2
U
2
Chi-squared
distance
,
V
(row
χ
)
(
X
−
E
)
C
−
1
U
,
C
−
1
/
2
V
2
(col.
χ
)
R
−
1
/
2
U
,
C
−
1
/
2
V
R
−
1
/
2
(
X
−
E
)
C
−
1
/
2
Correlation
I
,
I
R
−
1
/
2
U
V
C
1
/
2
1
/
2
,
C
1
/
2
V
2
(case A)
1
/
R
−
1
R
1
/
2
,
C
−
1
/
2
R
−
1
/
2
U
Row profiles
(
X
−
E
)
R
−
1
/
2
U
,
C
1
/
2
V
(case B)
W
−
1
1
(
X
−
E
)
C
−
1
/
2
W
−
1
2
W
−
1
1
U
V
W
−
1
2
W
−
1
1
1
/
2
,
W
−
1
2
R
−
1
/
2
1
/
2
General
W
1
,
W
2
U
V
