Till now in first two sessions of Business analysis till
some of our major learning are:-
·
Basics of SPSS
·
Understanding various variable meaning (Nominal,
ordinal & scale) and category variable
·
How SPSS results can are used in various
research analysis
·
Understanding use of null hypothesis ,chi-square
in research analysis
·
Divisive clustering
·
Agglomerative clustering
·
Cluster Analysis – Hierarchy clustering( for
less than 50 objects/ variable) , K- mean clustering( for more than 50 variable
Day 3 we had learned enhanced our
learning of business analysis by more details analysis and various techniques
or driving thoughts that can be used to
support our analysis and hence leading to a more specified targeted marketing
mix to be applied on the product or service
In our learning process we analyzed an data of cell phone
users recorded ,
Total 44 variable were recorded described below and
Total no. of cases / Responses recorded are 206
·
To analyze any first we should make cluster
/group and then compare with the variable
Cluster :- Cluster
are basically what you want to group based
on usuage similarity eg.we can make group 206 cases on the bases of service
providers, pre-paid & post-paid , monthly expenditure, usage etc or any
attribute or variable
How to make cluster ( fig 2)
If the variable are less than 50 we choose hiearachical
cluster and if more than 50 we should chose K-Mean Cluster ,
After required
selection following window pop up
1) Select the required variables to be clustered in variable column and make
sure the variable is selected in the cluster
2) In Statistics , make sure we had selected “Agglomeration schedule
“ and “proximity matrix)
3) In Plot , select “Dendrogram”
4) in Method select binary if the variable group responses
are in binary form and “interval” ->”Euclidean distance” if the variable
linear and responses are values
Concept of
Dendrogram:
Dendrogram plays an very important role while analysis of
data
The dendrogram is a visual representation of the spot
correlation data. The individual spots are arranged along the bottom of the
dendrogram and referred to as leaf nodes. Spot clusters are formed by joining
individual spots or existing spot clusters with the join point referred to as a
node. This can be seen in the diagram above. At each dendrogram node we have a
right and left sub-branch of clustered spots. In the following discussion, spot
clusters can refer to a single spot of a group of spots. The vertical axis is
labelled distance and refers to a distance measure between spots or spot
clusters. The height of the node can be thought of as the distance value
between the right and left sub-branch clusters. The distance measure between
two clusters is calculated as follows:
D=1-C (where D =
Distance and C = correlation between spot clusters.)
What can we
do with this information?
Draw a dendrogram showing clusters of spots according to how
strongly correlated the spots are. This correlation can be seen in the
expression profiles of spots from the same cluster.
Cluster is defined by Proximity and jaccard index
We figure out the cut off line b/w the cluster , Cut off
line is the one where the increase in coefficient is maximum and
so with the help of cluster and cut off and proximity index
we found where to focus our marketing mix and what kind of offers can company
offer to customer
Rules while Data Analysis are:-
1.
We must Combine different data for different
variable first and try to figure out is their any relation in initial stage
using cross tab and frequency analysis
2.
Should try to combine different procedures &
intergrate 2 outputs & final common result& analysis
Concept of OLAP
CUBE:-
·
OLAP stands for online analysis processor
·
An OLAP cube is a method of storing data in a
multidimensional form, generally for reporting purposes. In OLAP cubes, data
(measures) are categorized by dimensions. OLAP cubes are often pre-summarized
across dimensions to drastically improve query time over relational databases.
The query language used to interact and perform tasks with OLAP cubes is
multidimensional expressions (MDX). The MDX language was originally developed
by Microsoft in the late 1990s, and has been adopted by many other vendors of
multidimensional databases.
·
it is a very rough method of finding out things
, as the result is in 3 axis and we can compare the result and drive analysis
via various dissection and slices
Before using OLAP
cube:-
Before using OLAP CUBE ,we should categories all the
required variable into
·
Group
variable :- Group variable are category variable , having regular interval or when value in data variable is having limited
value
·
Summary
Variable:- Summary variable are scale variable , through which we can derive
mean, medium, mode and direct values helps in analyzing and comparing
How to use OLAP clube
next
Select the summary variable and
group variable according to the category they call and then press “ok” to find
out the comparison analysis of the data
This procedure is also known an STORY FORMING
This ,which is the main purpose of doing all this analysis which which help in designing the right marketing mix and design our story so that the emphasis is directly on the desired target which help company intern to more revenue and growth
By:- Parveen Rathee
Team - F
No comments:
Post a Comment