CIS 370

Business Analytics
Chapter 1: Introduction to Business Analytics
Instructor: Hamed Qahri-Saremi, PhD
What is Business Analytics
• Business analytics is the data-oriented approach to business problem solving and decision making, at
the intersection of:
– Statistics
– Computer Science
– Information Systems

• Business analytics combines qualitative reasoning with quantitative tools.

– Identify key business problems
– Define business problems in form of data analytics questions
– Translate data analysis results into decisions
– Improve business performance

• Success at business analytics needs…

– Knowledge of the right tools and techniques.
– Critical thinking.
– Practice.
Why Business Analytics:
Data Perspective
• Increasing Data exhaust
(trace data):

• Trail of data left by the

activities of Internet or
other computer system
users during their online
activity, behavior, and
transactions.

• Source of decision
insights.
2023
Why Business Analytics:
Data Perspective

Total data created worldwide Metric System for Measuring Data

from 2010 to 2025 Volume
(in zetta bytes)

https://en.wikipedia.org/wiki/Metric_prefix
https://www.statista.com/statistics/871513/worldwide-data-created/
 Why Business Analytics:
Technology Perspective – Evolution of Computing Technology
• Increasingly Expanding Software Solutions Market:

• The global big data & business analytics

market size was valued at USD 271.83
Billion in 2022.

• It is projected to reach USD 745.15 Billion by

2030, growing at a compound annual growth
rate (CAGR) of 13.5% from 2022 to 2030.

https://www.wsj.com/articles/huangs-law-is-the-new-moores-law-and-e
xplains-why-nvidia-wants-arm-11600488001
Why Business Analytics:
Decision Making Perspective

• Yet, we are bounded in our cognitive capacity!

– We have “Bounded Rationality” (Coined by Nobel Laurette, Herbert Simon, in 1957).
• The idea that when individuals make decisions, their rationality is limited by the tractability of the decision
problem, the limitations of their short-term memory, and the time available to make the decision.
– Decision-makers in this view act as satisficers; seeking a satisfactory solution rather than an optimal one.

• “Necessity is the mother of invention”.

– Solutions:
• Business Analytics
• Data Science
How Business Analytics
• Business analytics begins with understating the business context.
1. Ask the right questions.
2. Identify and execute the appropriate analyses.
3. Communicate (report and/or visualize) the results.
How Business Analytics
• Numerical results are not very useful unless they are accompanied with clearly stated
actionable business insights.

• There are three different types of analytics techniques.

– Descriptive analytics:
• What has happened?
– Describing what we know.

– Predictive analytics:
• What will happen in the future?
– Predicting unknowns based on knowns.

– Prescriptive analytics:
• What should we do?
– Scenario analysis.
– Turning data-driven recommendations into action also requires thoughtful consideration and organizational
commitment beyond developing descriptive and predictive analytical models.
Descriptive Analytics
• Descriptive Analytics: Describing what we know.
– Gather (data)
– Organize
– Tabulate
– Visualize
– Summarize

• Descriptive information can be presented in several formats (e.g., written reports, tables,
graphs, maps), such as:
– A firm’s marketing expenses and sales
– Financial reports
– Crime rates across regions and time
– COVID Cases: https://coronavirus.jhu.edu/us-map

• Used at the beginning and/or at the end of the analytics process.

Predictive Analytics
• Predictive Analytics: predicting unknowns based on knowns.
– Use historical data to make predictions
– Analytical models to help identify associations
– Associations used to estimate the likelihood of a favorable outcome
– Build models that help an organization understand what might (probabilistically) happen in the future
– Use statistics and data mining

• Examples
– Identifying customers who are most likely to respond to specific marketing campaigns
– Transactions that are likely to be fraudulent
– Incidence of crime at certain regions and times
– Products that customers are more likely to purchase
Prescriptive Analytics
• Prescriptive Analytics: what scenario/decision/action is the best?
– Optimization and simulation algorithms to provide advice
– Explore several possible actions to find an optimal solution
• Scenario analysis
– Suggest course of action

• Examples
– Scheduling employees’ works hours
– Business Process Simulation and Improvement
– Select a mix of products to manufacture
– Choose an investment portfolio
Business Analytics
Let’s Try…
Example Descriptive? Predictive? Prescriptive?
Which candidate will win the election? X
What price for a product will maximize profit? X
How much money do I need to save each year to have
enough money for retirement? X
How many products were sold last year? X
What’s the best route for the delivery person to drop off
packages to minimize the time needed to deliver all the X
packages?
How will marketing affect the daily sales of a product?
X
What was the average purchase price for new customers last
year? X

Other real-
world
examples?
Types of Data
• An important first step for making decisions is to find the right data and prepare it for
analysis.
– Compilation of facts, figures, or other content
– Numerical and non-numerical
– All types and formats are generated from multiple sources
– Often, we have a large amount of data
– Even small data can give insights
Types of Data
• It is not feasible to collect data that comprise the whole population of all elements of
interest.
– Too expensive
– It is impossible
– Traditional statistical techniques use sample information to draw conclusions about the population.

• A sample is a subset of the population and is used for analyses.

– Examples: election polls, customer surveys
Types of Data
• Sample data is generally collected in two ways:
– Cross-sectional data
• Collected by recording a characteristic of many subjects at the same point in time.

– Time series data

• Collected over several time periods focusing on certain groups of people, specific events,
or objects.
– Hourly, daily, weekly, monthly, quarterly, or annual observations.
Types of Data:
Cross-Sectional
• Cross-sectional data – all
data are static. Variance is
across records.
– NBA teams' standings in the
Western Conference at the end
of the 2022-2023 season.
Types of Data:
Time-Series
• Time series data –
temporal ordering is
relevant and meaningful.
– Unemployment rate in the US.
Types of Data
• Structured data
– Reside in a pre-defined, row-column format
• Imagine a two-dimensional table format
– Spreadsheet or database applications
– Enter, store, query, and analyze
– Numerical information that is objective and not open to interpretation

• Unstructured data
– Do not conform to a pre-defined, row-column format
– Textual
– Multimedia content
– Do not conform to traditional relational database structures

• Historically, companies relied mostly on structured data because of

– High cost to store and process unstructured data
– Performance limitations
Variables & Scales of Measurement
• A variable is a characteristic of interest that differs in kind or degree among various
observations (records).
• There are two types of variables:

Categorical Numerical

 Also called qualitative.  Also called quantitative.

 Represent categories.  Represent meaningful numbers.
 Labels or names to identify distinguishing  Arithmetic operations are meaningful.
characteristics.  Discrete: assumes a countable number of
 Arithmetic operations on the labels/values values (Example: number of children in a
are not meaningful. family).
 Continuous: assumes an uncountable
 Coded into numbers for data processing.
number of values within an interval (Example:
 Example: marital status. investment returns).
Variables & Scales of Measurement
• Analysis techniques depend on the type of data.
• There are four major scales: Star-rating
Gender in Reviews Temp. Profit
(1-5)
Scales of Measurement
• The Nominal Scale
– The least sophisticated level of measurement.
– Data are simply categories for grouping the data.
– Nominal variables have values with no logical ordering.

Qualitative values may be converted

to quantitative values for

analysis purposes.
Scales of Measurement
• The Ordinal Scale
– Ordinal data may be categorized and ranked with respect to some characteristic or trait.
• For example, instructors are often evaluated on an ordinal scale (excellent, good, fair, poor).

– Differences between categories are meaningless because the actual numbers used may be arbitrary.
• There is no objective way to interpret the difference between instructor quality.
Scales of Measurement
• The Interval Scale
– Data may be categorized and ranked with respect to some characteristic or trait.
– Differences between interval values are meaningful.
• Thus, the arithmetic operations of addition and subtraction are meaningful.
– No “absolute 0” or starting point defined.
• Meaningful ratios may not be obtained (What does mean?)
• For example, consider the Fahrenheit scale of temperature.
– This scale is interval because the data are ranked and differences (+ or −) may be obtained.
– But there is no “absolute 0” (What does 0°F mean?)
– There is one exception:
» The temperature scale in Kelvin is a ratio scale variable because its zero
value is absolute zero (i.e., nothing can be measured at a lower temperature
than 0 degrees Kelvin. No energy from molecular motion (that is, heat
energy) is available for transfer to other systems).
Scales of Measurement
• The Ratio Scale
– The strongest level of measurement.
– Ratio data may be categorized and ranked with respect to some characteristic or trait.
– Differences between interval values are meaningful.
– There is an “absolute 0” or defined starting point. “0” does mean “the absence of …” Thus, meaningful ratios
may be obtained.
– The following variables are measured on a ratio scale:
• General Examples: Weight, Time, and Distance.
• Business Examples: Sales, Profits, and Inventory Levels.
Variables & Scales of Measurement
Example 1.3: The owner of a ski resort gathers data from twenty tweens.

Nominal Ordinal Interval Ratio

– Music: On your car drive to the resort, which music streaming service was playing?
– Food quality: Q2: On a scale of 1 to 4, rate the quality of the food at the resort (where 1 is poor, 2 is
fair, 3 is good, and 4 is excellent.)
– Closing time: Presently, the main dining area closes at 3:00 pm. What time do you think it should
close?
– Own money spent: How much of your own money did you spend at the lodge today?
Data File Formats
• There are standard file formats.
– Fixed-width format: each column starts and ends in the same place in every row (.txt file).

– Delimited format: a delimiter separates fields, typically a comma (.csv file), or a tab (.tsv file).

– Excel format.

– Extensible Markup Language (XML): structured data, each piece enclosed in a pair of tags, gives
information on what the data are.

– HyperText Markup Language (HTML): structured data with tags, gives information on how to
display the data.

– JavaScript Object Notation (JSON): an alternative to XML, transmit human-readable data in

compact files, not as verbose as XML, supports wide range of data types, parsing is faster .