Documenti di Didattica
Documenti di Professioni
Documenti di Cultura
ASSISTANT PROFESSOR
DEPARTMENT OF MANAGEMENT
FACULTY OF SOCIAL SCIENCES
DEI
ARVIND BANGER
YES
NO
Dependence
methods
Interdependence
methods
ONE
Several
Is it metric?
YES
NO
Multiple
Regression
Multiple
discriminant
analysis
YES
YES
Factor
analysis
NO
Cluster
analysis
Metric
MDS
NO
Non-metric
MDS
Latent structure
analysis
Canonical analysis
MAV
ARVIND BANGER
ARVIND BANGER
DUMMY VARIABLE
ARVIND BANGER
IMPORTANT MULTIVARIATE
TECHNIQUES
ARVIND BANGER
ARVIND BANGER
Contd
E.g. if an individual is 20 years old, has income of Rs 12,000
Z=b1(20)+b2(12000)+b3(10)
ARVIND BANGER
FACTOR ANALYSIS
APPLICABLE WHEN there is systematic
ARVIND BANGER
10
i.e.
a large set of measured variables is resolved into
a few categories called
factors
ARVIND BANGER
11
objects
a1
b1
c1
k1
a2
b2
c2
k2
a3
b3
c3
k3
aN
bN
cN
kN
ARVIND BANGER
12
13
Contd
EIGEN VALUE : It is the sum of squared values of the factor
ARVIND BANGER
14
ARVIND BANGER
15
CENTROID METHOD
This method tends to maximize the sum of loadings
ILLUSTRATION :
Given the following correlation matrix, R, relating to 8
variables with unities in the diagonal spaces. Work out
the first & second centroid factors:
ARVIND BANGER
16
Variables
1
Variables
0.709
0.204 0.051
0.22
ARVIND BANGER
0.12
0.891
0.12
0.152
0.152
17
SOLUTION:
a)
b)
c)
ARVIND BANGER
18
Variables
1
Variables
0.709
0.204 0.051
0.22
column sum(Si)
0.891
0.12
0.152
0.152
19
0.693
0.618
0.642
0.641
0.629
0.694
0.679
ARVIND BANGER
0.683
20
ARVIND BANGER
21
0.693
0.48
0.42
0.422
0.43
0.42
0.43
ARVIND BANGER
23
0.52
0.281 -0.24
-0.36
0.19
-0.37
-0.32 0.301
-0.34
0.23
3 -0.24
4 -0.36
0.192 -0.28
0.19
0.43
0.146 -0.37
-0.22 0.294
6 -0.37
0.33
7 -0.32
0.33
0.539 -0.31
-0.31 0.534
8 0.301 -0.23
-0.37
ARVIND BANGER
-0.35
24
ARVIND BANGER
25
Variables
3*
4*
0.19
6*
7*
0.52
0.43
0.23
0.146 0.366
0.19
0.39
0.217 0.294
0.39
0.518
0.33
0.33
0.539 0.312
0.354
0.301
0.23
2.58
2.642
2.47
26
Factor loadings
Centroid factor A
Centroid factor B
0.693
0.563
0.618
0.577
0.642
-0.539
0.641
-0.602
0.629
0.558
0.694
-0.63
0.678
-0.518
0.683
0.593
ARVIND BANGER
27
Commonality(h2)
Factor loadings
Centroid factor A
Centroid factor B
A2+B2
0.693
0.563
0.797
0.618
0.577
0.715
0.642
-0.539
0.703
0.641
-0.602
0.773
0.629
0.558
0.707
0.694
-0.63
0.879
0.678
-0.518
0.729
0.683
0.593
0.818
ARVIND BANGER
28
Proportion of variance
Variables
Factor loadings
Commonality(h2)
Centroid factor A
Centroid factor B
Eigen value
3.49
2.631
6.121
Proportion of
0.44
0.33
0.77
total variance
[44%]
[33%]
[77%]
Proportion of
0.57
0.43
common variance
[57%]
[43%]
[100%]
ARVIND BANGER
29
PRINCIPAL-COMPONENTS METHOD
This method seeks to maximize the sum of squared loadings
of each factor .Hence the factors in this method explain more
variance than the loadings obtained from any other method of
factoring.
Principal components are constructed which are linear
combination of given set of variables.
p1 = a11X1+a12X2+.+a1nXn
p2= a21X1+a22X2+.+a2nXn
and so on till pn
The aijs are called loadings and worked out in such a way that
PC are uncorrelated(orthogonal) and first PC has maximum
variance.
ARVIND BANGER
30
ILLUSTRATION:
Take the correlation matrix R for 8 variables and compute:
(i) the first two principal component factors.
(ii) the communality for each variable on the basis of said
two component factors.
(iii) the proportion of total variance as well as the proportion
of common variance explained by each of the two
component factors.
ARVIND BANGER
31
SOLUTION:
As the correlation matrix is positive manifold we work out the
1st principal component factor as follows:
The vector of column sums is referred to as Ua1 and when it
ARVIND BANGER
32
Variables
1
Variables
0.709
0.204 0.051
0.22
0.12
0.891
0.12
0.152
0.152
Va1
=
Ua1/normalizing
factor
33
Normalizing factor:
={(3.622)2+(3.263)2+(3.392)2+(3.385)2+(3.324)2+(3.666)2+(3.587)2+
(3.605)2}
=9.868
ARVIND BANGER
34
Variables
(Characteristic
vector Va1)
normalizing
factor
Principal
of Ua2
component
0.371
1.868
0.69
0.331
1.868
0.62
0.334
1.868
0.64
0.343
1.868
0.64
0.337
1.868
0.63
0.372
1.868
0.70
0.363
1.868
0.68
0.365
1.868
0.68
ARVIND BANGER
35
Principal component
II
1
0.57
0.59
-0.52
-0.59
0.57
-0.61
-0.49
-0.61
ARVIND BANGER
36
Variables
Principal components
commonality(h2)
II
I2+II2
0.69
0.57
0.801
0.62
59
0.733
0.64
-0.52
0.68
0.64
-0.59
0.758
0.63
0.57
0.722
0.7
-61
0.862
0.68
-0.49
0.703
0.68
-0.61
0.835
Eigen value
3.4914
2.6007
6.0921
Proportion of
0.436
0.325
0.761
total variance
43.6%
32.5%
76%
Proportion of
0.537
0.427
1.00
57%
43%
100%
common varianceARVIND
BANGER
37
ARVIND BANGER
38
CLUSTER ANALYSIS
Unlike techniques for analyzing relationship between
STEPS:
Selection of sample to be clustered (buyers, products etc.).
Definition of the variables on which to measure the
objects, events etc. (e.g. market segment characteristics,
product competition definitions etc.)
Computation of similarities among entities (through
correlation, euclidean distances, and other techniques)
Selection of mutually exclusive clusters(maximisation of
within cluster similarity and between cluster differences).
Cluster comparison & validation.
ARVIND BANGER
39
A minivan buyers
B sports car buyers
Income
Age
Family size
ARVIND BANGER
40
Method:
41
CONJOINT ANALYSIS
Used in market research & product development
42
ARVIND BANGER
43
THANK YOU
ARVIND BANGER
44