Analytical Decision Making

Prepared By Deepali Sonawane, Assistant Professor, SIOM

Introduction of Analytical Decision Making
 Analytical decision making is a process of carefully analyzing a data and come
up with a solution.
 It is the actual selection from among alternatives of a course of action.
 It is at the core of planning.
 It is not a easy job, it require analytical skills.
 It is affected by number of factors.
 A good decision is taken by adopting a procedure.
 This process is depending on nature of the problem and the nature of the
organization.

Prepared By Deepali Sonawane, Assistant Professor, SIOM

Analytical Decision Maker
 Analytical decision makers carefully analyze a data and come up with a solution.
 They are careful and adaptable thinkers.
 They will invest time to clean information to form a conclusion.
 These decision-makers are task-oriented.
 Analytical decision-makers take time to compile data and evidence before they
come to a conclusion.
 When they do make a decision, they have looked at all the details and formed what
they believe is the best possible solution.

Prepared By Deepali Sonawane, Assistant Professor, SIOM

Analytical Decision Making Process
1. Identification of a problem
2. Diagnosing the problem
3. Collect and analyze the relevant information
4. Discovery of alternative course of action
5. Analyzing the alternatives
6. Selection of best alternative
7. Conversion of decision into action
8. Verifying the decision

Prepared By Deepali Sonawane, Assistant Professor, SIOM

1. Identification of a problem
 Recognizing a problem
 Problem arises due to difference between what is and what should be.
 Changes in the business environment are the main reason for creating a
problem.
 A well defined problem is half solved.
 Example : If staff working with unavailability of internet. So internet is the
problem.

Prepared By Deepali Sonawane, Assistant Professor, SIOM

2. Diagnosing the problem
 There is a difference between identifying the problem and diagnosing the
problem.
 Example : A doctor can diagnosed the patient illness. The patient can not find
what the real disease is. But doctor can do so with the help of information given
by the patient.
 In management, the manager acts as a doctor when diagnosing the problem.

Prepared By Deepali Sonawane, Assistant Professor, SIOM

3. Collect and analyze the relevant information
 Analytical Decision maker should collect the information and study it
carefully to analyze the problem.
 If problem is analyzed, a quick and quality decision made by the decision
maker.
 He must verify that only relevant information is collected and analyzed.
 Example: If you want a marketing consultancy and you are considering a pay per
click advertising campaign, you may want to have information such as
which keywords customer use most when searching for results related to
consulting.

Prepared By Deepali Sonawane, Assistant Professor, SIOM

4. Discovery of alternative course of action
 Creative thinking is necessary to develop or discover many alternative courses of
action.
 If there is no alternative, there is no need to make a decision.
 If there are many alternative, the decision maker will have more freedom to
make decisions.
 Example : A Person believe that he must have a car to go to work. However, he
should consider other alternatives, such as public transportation, renting a car,
shared car ownership or a company car.

Prepared By Deepali Sonawane, Assistant Professor, SIOM

5. Analyzing the alternatives
 The generation of alternatives must be followed by a through analysis of the
pros and cons of each alternative.
 The alternatives must be evaluated to see how effective each would be.
 The decision maker can prepare a list of limits for each alternative.
 Example : Consider two alternatives that are equally risky, but one will cost more
and the other will take longer to implement. In this case, the decision would
depend on whether cost or time is more important.

Prepared By Deepali Sonawane, Assistant Professor, SIOM

6. Selection of best alternative
 After careful evaluation, the decision maker can select the best alternative.
 An alternative that gives maximum benefits to the organization is selected
and also it must be suit organization objectives.
 Example : Problem is which car to buy? What is important to us that will help
us to determine which car best suit our situation. Customer has to consider all
parameters like style, comfort, noise, fuel consumption, speed, price, available
payment terms, reliability etc.

Prepared By Deepali Sonawane, Assistant Professor, SIOM

7. Conversion of decision into action
 Based on all thinking, analysis and evaluation, now is the time to implement
the decision.
 Whether the decision you have made is correct or not, it will only be
determined if you implement it.
 The implementation of the decision requires constant monitoring so that the
expected result of the optimal course of action are obtained.
 Example : The upgrade of infrastructure of an organization must be
planned in a series of stages that in total span a period of months or years.

Prepared By Deepali Sonawane, Assistant Professor, SIOM

8. Verifying the decision
 Decision maker should monitor the decision implementation process to
make sure everything is processing according to plan.
 He must also ensure that the problem that started the decision making process
has been resolved.
 Example : We need to check that after starting Covid-19 vaccination program,
are we able to solve corona pandemic problem or not.

Prepared By Deepali Sonawane, Assistant Professor, SIOM

Characteristics of the Analytical Decision Making Process
 Selective: It is a selective process in which the optimal alternative is opted,
among the various alternatives. The selection of the alternative is done, only after
evaluating all the alternatives against the objectives.
 Mental and intellectual process: : It is a mental and intellectual process
because whatever decisions are taken, they are based on logical study to make
them more effective. For which intelligence, knowledge, experience, educational
level, and mental facilities are essential.
 Dynamic: It is a dynamic activity in the sense that a particular problem may have
different solutions, depending upon the time and circumstances.
 Positive or Negative: A decision is not always positive, sometimes even after
analyzing all the points a decision may turn out as a negative one.
 Goal oriented: Decisions should be goal oriented for the organization.
Prepared By Deepali Sonawane, Assistant Professor, SIOM
Characteristics of the Analytical Decision Making Process
 Ongoing process: We all know that in a company various decisions are
taken daily by different levels of management to keep the firm going. These
decisions are taken by keeping in mind the objectives of the organization.
 Evaluative: Evaluation of the possible alternatives using critical appraisal
methods, is a part of the decision-making process.
 Choice of alternative: Decision maker has freedom to choose an
alternative.
 Measurement of Performance: Decision making is a measurement on the
basis of which the success or failure and execution or non-execution of the
decisions taken by the managers.

Prepared By Deepali Sonawane, Assistant Professor, SIOM

Breaking down a business problem into key questions
1. What business problem do we want to solve? (in a precise, well defined
way)
2. What exactly will be the commercial benefit of solving the problem?
3. What data is available that might help to solve the problem?
4. How exactly can we use data to solve the problem?
5. How much pre-processing will be required?
6. What techniques and technologies will be used?
7. What timelines, resources and budgets are required?
8. Do the plans fit in with the rest of the business?
9. Are they aligned with your overall data science strategy?
Prepared By Deepali Sonawane, Assistant Professor, SIOM
Characteristics of good questions
 Relevant
 Clear
 Concise
 Purposeful
 Guiding to get alternatives
 Stimulates Thinking / Encouragement for thinking
 Single-Dimensional
 Define aim / objectives
 Brief and direct

Prepared By Deepali Sonawane, Assistant Professor, SIOM

Business analyst
 Business analysts are responsible for using data to inform strategic business
decisions.
 Business analysts tend to be more involved in addressing business needs and
recommending solutions.
 Duty of Business Analyst is evaluating business processes for efficiency, cost,
and other valuable metrics.
 Presenting strategic recommendations for process adjustments,
procedures, and performance improvements.
 Creating visuals and financial models to support business decisions.

Prepared By Deepali Sonawane, Assistant Professor, SIOM

Skills of a good business analyst
 Understanding the Business Objective
 Analytical and Critical Thinking
 Communication and Interpersonal Skills
 Negotiation and Cost-Benefit Analysis
 Decision-Making Skills
 Some knowledge of Programming Languages
 Creation of Reports
 Some knowledge of Database and SQL
 Knowledge of statistics and probability
 Problem Solving technique
 Documentation

Prepared By Deepali Sonawane, Assistant Professor, SIOM

Business analytics applications
1. Marketing Analytics
2. HR Analytics
3. Supply Chain Analytics
4. Finance Analytics
5. Retail Industry Analytics
6. Sales Analytics
7. Web & Social Media Analytics
8. Healthcare Industry
9. Energy Analytics
10. Transportation Analytics
11. Lending Analytics
12. Sports Analytics

Prepared By Deepali Sonawane, Assistant Professor, SIOM

Basic Tools of Business Analytics
 Business analytics tools are types of application software that
retrieve data from one or more business systems and combine it in a
repository, such as a data warehouse, to be reviewed and analyzed.
 Following are the most commonly used basic tools of business analytics:
➢MS Excel
➢Tableau
➢Power BI

Prepared By Deepali Sonawane, Assistant Professor, SIOM

Statistical Analysis
 Statistics is the science of collecting, organizing, presenting, analyzing
and interpreting data to help in making more effective decisions.
 Statistical analysis is the process of collecting large volumes of data
and using statistics and other data analysis techniques to identify
trends, patterns, and insights.
 Statistical analysts take raw data and find correlations between
variables to reveal patterns and trends to relevant stakeholders.
 Statistical analysis is a scientific tool in AI and ML that helps collect and
analyze large amounts of data to identify common patterns and trends to
convert them into meaningful information.
Prepared By Deepali Sonawane, Assistant Professor, SIOM
Types of Statistical Analysis
1. Descriptive Statistical Analysis: Descriptive statistics consists of
organizing and summarizing the data without conclusions about its
contents. Example: Business gave book of its expenses and analyst
summarized the percentage of money it spent on different categories of items.
2. Inferential Statistical Analysis: Inferential statistics takes the results of
descriptive statistics and drawing conclusions from the data and making
recommendations. It is often used by businesses to inform company decisions
and in scientific research to find new relationships between variables.
Example: Instead of only summarizing the business’s expenses, analyst might
go on to recommend in which areas to reduce spending and suggest an
alternate budget.

Prepared By Deepali Sonawane, Assistant Professor, SIOM

Statistical Analysis Methods
Descriptive Analytics Inferential Analytics
1. Mean 1. Hypothesis Testing
2. Mode i. z-test
3. Median ii. t-test
iii. chi square test
4. Variance
iv. ANOVA
5. Standard Deviation
6. Regression

Prepared By Deepali Sonawane, Assistant Professor, SIOM

Measuring Central Tendencies and Dispersion

Prepared By Deepali Sonawane, Assistant Professor, SIOM

 Regression
 Regression: It is statistical method that helps to understand and predict the
relationship between the dependent and independent variable.
 It describes how one variable (dependent variable) changes as another variable
(independent variable) changes.
 Example: Predicting salary based on year of experience, Predicting exam score
based on study hours, predicting resale price based on vehicle usage.

Prepared By Deepali Sonawane, Assistant Professor, SIOM

Hypothesis Testing
1. Hypothesis is tentative, intelligent guesses as to the solution of the
problem.
2. Hypothesis is a specific statement of prediction.
3. It describes in concrete terms what you expect to happen in the
study.
4. Hypothesis is an assumption about the relations between the variables.
5. Hypothesis is a predictive statement that relates an independent
variable to a dependent variable.
6. Hypothesis must contain at least one independent variable and one
dependent variable.

Prepared By Deepali Sonawane, Assistant Professor, SIOM

Types of Hypothesis
1. Null Hypothesis: In case of Null Hypothesis, the statements assert that
there is no relationship between variables. This hypothesis are also known
as Statistical Hypothesis as they can be tested. The researcher needs to
prove the null hypothesis wrong by showing that the relationship between the
variables do exist. It is denoted as H0 or HO. Example: There is no
relationship between family’s income and expenditure on recreation.
2. Alternative Hypothesis: An alternative hypothesis states that there is a
relationship between the existing variables. The alternative hypothesis
should be supported by an empirical evidence to prove the null hypothesis
wrong. It is denoted by H1 or Ha. Example: There is a positive
relationship between family’s income and expenditure on recreation.

Prepared By Deepali Sonawane, Assistant Professor, SIOM

Key terms of Hypothesis Testing
 Level of Significance: It refers to probability in which we accept or reject
the null hypothesis.
 Degree of freedom: The number of elements in the calculation which are
able to vary is known as degree of freedom.
 P-Value: The P-value is the probability of finding the observed results when
the null hypothesis of a study given problem is true.
 Critical Value: It is threshold or cutoff point used to determine whether
to reject the null hypothesis in a hypothesis test.
 One-sided Vs. Two sided testing: One sided tests suggest the possibility of
an effect in single direction only. Two sided tests suggest two directions-Positive
and negative.
 Sampling: Statistical analysts can test a hypothesis by examining and measuring
a random sample of the population that is being examined.
Prepared By Deepali Sonawane, Assistant Professor, SIOM
Errors in Hypothesis Testing
 Type I Error: If we reject correct hypothesis then it comes under Type I
error and denoted by α. Probability of Type I error(α) = probability of
rejection H0 when H0 is true. When the difference of two means is zero
and test indicates the rejection of the null hypothesis, then it is known as
Type I error.
 Types II Error: If we accept the wrong hypothesis then it comes under
Type II error and denoted by β. Probability of Type II error(β) =
probability of accepting H0 when H0 is false i.e. H1 is true. When the
difference of two means is not zero and test indicates the acceptance of
the null hypothesis, then it is known as Type II error.

Prepared By Deepali Sonawane, Assistant Professor, SIOM

Errors in Hypothesis Testing

Prepared By Deepali Sonawane, Assistant Professor, SIOM

Data Visualization
 Data visualization is the graphical representation of data by using visual
elements like charts, graphs, and maps.
 Data visualization tools provide an accessible way to see and understand
trends, outliers, and patterns in data.
 The primary goal of data visualization is to make data more accessible and
easier to interpret, allowing users to identify patterns, trends, and
outliers quickly.
 This can include a variety of visual tools such as:
• Charts: Bar charts, line charts, pie charts, etc.
• Graphs: Scatter plots, bubble plots, histograms, etc.
• Maps: Geographic maps, heat maps, etc.
• Dashboards: Interactive platforms that combine multiple visualizations.
Prepared By Deepali Sonawane, Assistant Professor, SIOM
Need of Data Visualization
 Communicating Insights
 Identifying patterns and trends
 Comparing Data
 Identifying Anomalies
 Making Predictions

Prepared By Deepali Sonawane, Assistant Professor, SIOM

Exploratory Data Analysis (EDA)
 Exploratory Data Analysis (EDA) is a data analysis method that involves using
visualization and transformation to understand a data set.
 Exploratory Data Analysis refers to the process of performing initial
investigations on data to discover patterns, to identify anomalies, to test
hypothesis and to check the assumptions with the help of summery statistics
and graphical representation.
 EDA is an iterative process that involves generating questions about the data,
visualizing, transforming, and modeling data, and using results to refine
questions and generate new ones.
 Python and R are the Exploratory Data Analysis tools.

Prepared By Deepali Sonawane, Assistant Professor, SIOM

Why Exploratory Data Analysis is important?
 Understanding Data Structures
 Identifying Patterns and Relationships
 Detecting Anomalies and Outliers
 Testing Assumptions
 Informing Feature Selection
 Optimizing Model Design
 Facilitating Data Cleaning
 Enhancing Communication

Prepared By Deepali Sonawane, Assistant Professor, SIOM

Types of Exploratory Data Analysis
 Univariate Analysis: Univariate analysis focuses on a single variable to
understand its internal structure. It is primarily concerned with describing
the data and finding patterns existing in a single feature. Graphics includes
Histogram, Box plots, Bar charts.
 Bivariate Analysis: Bivariate evaluation involves exploring the connection
between variables. It enables find associations, correlations, and
dependencies between pairs of variables. Graphics includes Scatter plots,
Correlation Coefficient, line graphs.
 Multivariate Analysis: Multivariate analysis examines the relationships
between two or more variables in the dataset. It aims to understand how
variables interact with one another, which is crucial for most statistical
modeling techniques. Graphics includes Pair Plots, Principal Component
Analysis.
Prepared By Deepali Sonawane, Assistant Professor, SIOM
Steps involved in Exploratory Data Analysis

Prepared By Deepali Sonawane, Assistant Professor, SIOM

Data Cleaning
 Data cleaning is the process of fixing or removing incorrect,
corrupted, incorrectly formatted, duplicate, or incomplete data within a
dataset.
 When combining multiple data sources, there are many opportunities
for data to be duplicated or mislabeled.

Prepared By Deepali Sonawane, Assistant Professor, SIOM

Data Inspection
 Data inspection is the act of viewing data for verification and debugging
purposes, before, during, or after a translation.
 Data inspection is a data mining technique that involves transforming
raw data into understandable format.
 It includes normalization and standardization, transformation, feature
extraction and selection etc.
 To ensure that we are dealing with the right information, we need a
clear view of our data at every stage of the transformation process.
 It involves examining and understanding data to ensure its accuracy,
consistency and relevance before conducting any further analysis
Prepared By Deepali Sonawane, Assistant Professor, SIOM
 Thank you

Prepared By Deepali Sonawane, Assistant Professor, SIOM