2. DESCRIPTIVE ANALYSIS
It's about finding what has happened by summarizing the data theiugh
Innovative methods and analysing the data through simple queries. Eg:
difference in shopping behaviour of men and women
It involves data summarisation using techniques such as pivot tables, descriptive
statistics and data visualisation.
It includes data summarisation. It includes descriptive statistics such as
measures of central tendency, measures of variation and measures of shape
which can provide useful insights.
Plots such as histogram, bar chart, pie chart, box plot, scatter plot and tree
diagram provide powerful insights about past data and helps to analyse further.
It is an important part of reporting across several industries which enables top
management to monitor key performance indicators and take decisions. Eg, a
retailer such as Big Bazar would like to know top 5 products sold by region, city,
store etc.
It will help the management to plan thier inventory, shelf space, pricing etc.
Total revenue generated over the last few years can also be monitored at
regional, city or store levels.
3. DATA IS CLASSIFIED BASED ON:
Application of descriptive analysis is designing effective dashboards
and scorecards. Data types and scales
Structure
Scale of measurement
Structure
Structured and unstructured
Structured: represented in the form of rows and columns. Eg,
employee data, student dataCannot be represented in the form of
rows and columns, unstructured data. Eg, emails, images generated
by medical devices, MRI, ECG, data on social media platforms, IoT
4. TYPE OF DATA COLLECTED:
Cross sectional, time series, panel data
Cross sectional: analysing data on multiple variables at the same
time. Eg, analysis of budget, directors, actors of the movie during
2017
Time series: single variable over several time periods. Eg, demand for
smartphones collected monthly, weekly, yearly, etc.
Panel data: several variables over several time intervals. Eg,
unemployment rates for several countries over several years.
5. TYPES OF DATA MEASUREMENT
SCALES
Nominal Scale: variables that are basically names and are also known
as categorical variables. Like for variable marital status, a data
collector may use 1 for single, 2 for married, 3 for divorced.
Ordinal scale: It is finite, usually a 5 point scale. Like for eg, 1 is poor,
2 is fair, 3 is good, 4 is very good, 5 is excellent.
Interval Scale: corresponds to a variable in which the value is chosen
from an interval set. Eg, temperature measured in centigrade.
Ratio Scale: any variable for which the ratios can be computed. Eg,
salary, marks. If A's salary is 20000 and B's salary is 40000 a month.
It can be calculated that B earns double than A.
6. POPULATION AND SAMPLE
Population is the set of all possible data for a given context.
Sample is the subset taken from the population.
For eg, while election, it is not possible to collect data from around
800 million eligible voters, rather it is done by opinion expressed by
a subset of voters called sample.