Scribd
Upload
46 views2 pages

Cluster Analysis. Discriminant Analysis. MDS

Uploaded by

Soumi Dey
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as DOCX, PDF, TXT or read online on Scribd
46 views2 pages

Cluster Analysis. Discriminant Analysis. MDS

Uploaded by

Soumi Dey
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as DOCX, PDF, TXT or read online on Scribd
You are on page 1/ 2

Cluster analysis

Cluster analysis is a descriptive analysis technique that groups object so that each object is similar to
the other in the cluster and different from the objects of the other cluster. It is a quantitative form of
classification. Therefore cluster analysis classifies variables into clusters. The variables in one cluster
have high correlation with one another and low correlation with variables in other clusters.

The goal is to determine the number of mutually and exhaustive clusters based on the similarities of
profiles among entities present in the population. The classification procedures used in cluster
analysis are based on either density of population or distance between members.

There are three major type of clustering

1. Hierarchical Clustering

2. Partitional Clustering

3. Density based Clustering

Assumptions-

1. The sample should be representative of the population

2. The variables should not be correlated. Even if variables are correlated remove correlated
variables or use distance measures that compensate for the correlation.

Multidimensional scaling

Multidimensional scaling or MDS is a set of procedures for representing perceptual or affective


dimensions of substantive interest. It provides useful methodology for portraying subjective
judgements of diverse kinds (Green, 1978).

Multidimensional scaling helps to measure and item in more than one dimension at a time. The basic
assumption is that a set of objects is perceived to be more or less similar to one another on a
number of dimensions instead of only one (Sheth, 1971).

Multidimensional scaling is a visual representation of distances or dissimilarities between sets of


objects. “Objects” can be colors, faces, map coordinates, political persuasion, or any kind of real or
conceptual stimuli (Kruskal and Wish, 1978).

Approaches in MDS

1. Nonmetric methods: the input is generated by rank-ordering pairs of objects, are more flexible
because no specific type of relationship between the calculated distance and the similarity measure
is assumed.

2. Metric methods: it assumes that input as well as output is metric. Rather than assuming that only
the ordered relationships are preserved in the input data, it can assumed that the output preserves
the interval and ratio qualities of the input data.
Discriminant Analysis

Discriminant analysis is the appropriate statistical techniques when the dependent variable is a
categorical (nominal or nonmetric) variable and the independent variables are interval (metric)
variables. It is a technique to discriminate two or more mutually exclusive and exhaustive groups on
the basis of some explanatory variables.

When two classifications are involved, it is called two-group discriminant analysis. When three or
more classifications are involved, it is called multiple discriminant analysis (MDA).

You might also like