BA Test Material
BA Test Material
Data Analysis
Both terms are often used interchangeably, but they have distinct meanings in business
analytics.
Tools Used Excel, SQL, Power BI, Tableau Python, R, SAS, Big Data tools (Hadoop,
for visualization. Spark), AI/ML frameworks.
Key Takeaway
● Data Analysis is a part of Data Analytics but focuses only on examining historical data.
● Data Analytics is a broader discipline that includes predictive, prescriptive, and
AI-driven analytics for making future business decisions.
Types of Analytics in Business Analytics
There are four main types of analytics, each serving a different purpose in data-driven
decision-making:
🔹 Definition: Descriptive analysis helps summarize historical data to identify patterns and
trends. It provides insights into what has already happened using simple statistical measures.
🔹 Key Techniques:
● Measures of Central Tendency: Mean, Median, Mode
● Measures of Dispersion: Standard Deviation, Variance, Range
● Data Visualization: Bar Charts, Pie Charts, Line Graphs
🔹 Example:
📌 E-commerce – Amazon analyzes past sales data to see how revenue has changed over
time.
📌 Healthcare – A hospital examines patient admission rates over the past year.
📌 Marketing – A company tracks website traffic and customer engagement over a month.
🔹 Definition: Diagnostic analysis examines past data to understand the reasons behind
trends and events. It answers why something happened.
🔹 Key Techniques:
● Correlation Analysis – Finding relationships between variables.
● Regression Analysis – Understanding the impact of one factor on another.
● Drill-Down Analysis – Exploring deeper layers of data to pinpoint causes.
🔹 Example:
📌 Retail – If sales dropped in December, a company might analyze customer reviews,
competitor pricing, and marketing effectiveness to find out why.
📌 Finance – A bank investigates why loan default rates increased by analyzing customer
profiles.
📌 Healthcare – If a hospital sees a rise in heart disease cases, it may analyze patient
lifestyle habits to find contributing factors.
🔹 Definition: Predictive analysis uses historical data, statistical models, and machine
learning to make informed forecasts about future events.
🔹 Key Techniques:
● Machine Learning Algorithms – Decision Trees, Neural Networks.
● Time Series Analysis – ARIMA, Exponential Smoothing.
● Probability Models – Monte Carlo Simulations.
🔹 Example:
📌 Stock Market – Investment firms use machine learning to predict stock price trends.
📌 Healthcare – AI predicts disease outbreaks based on patient data trends.
📌 Retail – Amazon predicts which products will be in high demand next month using
customer purchase history.
🔹 Key Techniques:
● Optimization Models – Linear Programming, Genetic Algorithms.
● Simulation Models – Predicting different scenarios and outcomes.
● Decision Trees – Mapping out different decision paths.
🔹 Example:
📌 Ride-Sharing (Uber) – Uber uses prescriptive analytics to adjust pricing based on
demand and traffic conditions in real time.
📌 Supply Chain Management – Companies use analytics to optimize inventory levels and
reduce waste.
📌 Healthcare – AI recommends the best treatment plans for cancer patients based on
predictive modeling.
Summary Table
Type Question Example
Answered
Predictive What will happen? Forecasting customer demand for the next
Analytics quarter.
📊 Use Cases:
● Fraud Detection: AI models detect suspicious transactions in real time.
● Risk Management: Credit scoring models assess loan default risks.
● Algorithmic Trading: Machine learning optimizes stock trading strategies.
● Customer Segmentation: Banks analyze spending patterns to personalize offers.
🔹 Example: JPMorgan Chase uses predictive analytics to prevent credit card fraud.
📊 Use Cases:
● Customer Targeting: Analyzing consumer behavior to personalize ads.
● A/B Testing: Comparing two ad versions to optimize conversions.
● Sentiment Analysis: AI monitors customer reviews and social media sentiment.
● Churn Prediction: Identifying customers likely to stop using a service.
3. Healthcare
📊 Use Cases:
● Disease Prediction: AI detects early signs of diseases from medical records.
● Patient Readmission Forecasting: Hospitals predict which patients may return.
● Drug Discovery: Machine learning accelerates new medicine research.
● Operational Efficiency: Predicting hospital bed occupancy to optimize resources.
🔹 Example: IBM Watson assists doctors in diagnosing diseases based on medical data.
4. Retail & E-Commerce
📊 Use Cases:
● Inventory Optimization: Predicting product demand to avoid overstocking.
● Price Optimization: AI dynamically adjusts prices based on demand and competitor
trends.
● Recommendation Systems: Personalized product suggestions using past purchase
data.
● Customer Lifetime Value (CLV) Analysis: Identifying high-value customers.
📊 Use Cases:
● Route Optimization: GPS and AI minimize fuel consumption and delivery time.
● Demand Forecasting: Predicting future inventory needs to reduce waste.
● Warehouse Management: Robotics and AI improve storage efficiency.
● Risk Mitigation: Identifying supply chain disruptions early.
6. Manufacturing
📊 Use Cases:
● Predictive Maintenance: Sensors detect machine failures before they occur.
● Quality Control: AI detects product defects in real-time.
● Supply Chain Analytics: Ensuring timely raw material procurement.
● Lean Manufacturing: Data-driven waste reduction strategies.
🔹 Example: General Electric (GE) uses IoT and AI for predictive maintenance.
📊 Use Cases:
● Performance Analysis: AI tracks athlete performance for training improvements.
● Fan Engagement: Personalized content recommendations for viewers.
● Ticket Pricing Optimization: Dynamic pricing based on demand.
● Game Strategy Optimization: Teams use analytics for tactical advantages.
📊 Use Cases:
● Crime Prediction: AI forecasts high-risk areas for law enforcement.
● Economic Policy: Data analytics helps in tax collection and economic planning.
● Smart Cities: Optimizing traffic management, waste disposal, and energy use.
● Public Health Surveillance: Tracking disease outbreaks (e.g., COVID-19 trends).
Summary Table
Industry Key Applications
Government & Public Crime prediction, tax fraud detection, smart cities
Policy
Types of Data in Business Analytics
Data can be classified in multiple ways based on structure, nature, and source. The key
classifications are:
Quantitative Data Data that can be measured and Sales revenue, temperature,
(Numerical Data) expressed in numbers. height, age
🔹 Example:
● Quantitative: A company’s monthly revenue is ₹10,00,000.
● Qualitative: Customer feedback: "Excellent service!"
Nominal Data Categorical data with no inherent Gender (Male/Female), Blood Type
order. (A, B, AB, O)
Continuous Data that can take infinite values Temperature, weight, height
Data within a range.
🔹 Example:
● Nominal: Eye color (Brown, Blue, Green).
● Ordinal: Movie ratings (1 star, 2 stars, 3 stars, etc.).
● Discrete: Number of students in a class (50 students).
● Continuous: Height of a person (5.8 ft, 5.85 ft, etc.).
Structured Data Organized in a fixed format, easily Excel sheets, SQL databases
stored in databases. (customer names, sales figures)
Semi-Structured Partially organized but lacks a JSON, XML files, emails, social
Data strict structure. media posts
🔹 Example:
● Structured: An Excel file with employee IDs and salaries.
● Semi-Structured: A JSON file containing customer reviews.
● Unstructured: A collection of CCTV footage from a retail store.
🔹 Example:
● Primary Data: A company conducts a customer satisfaction survey.
● Secondary Data: A business uses World Bank reports for market research.
5. Based on Sensitivity (Personal, Sensitive, Public Data)
Type Description Example
Public Data Available for public use with no Government census reports, company
restrictions. stock prices
🔹 Example:
● Personal: A customer’s email address.
● Sensitive: A bank’s account login credentials.
● Public: India’s GDP data published by the government.
Summary Table
Category Types of Data Examples
Big Data refers to extremely large and complex datasets that cannot be managed, processed,
or analyzed using traditional data processing tools. It requires advanced technologies such as
machine learning, artificial intelligence (AI), cloud computing, and distributed databases
for efficient handling.
🔹 Example:
● Amazon processes over 1 million customer transactions per second.
● YouTube users upload 500+ hours of video every minute.
Velocity Speed at which data is created, collected, and Stock market data updates in
processed. real-time.
Variety Different types and formats of data (structured, Text, images, videos, IoT
semi-structured, unstructured). sensor data.
Structured Data Organized in fixed formats like SQL databases, Excel files.
databases.
Semi-Structured Partially organized but lacks strict JSON, XML, emails, social
Data formatting. media posts.
Unstructured Data No predefined format, difficult to Videos, images, PDFs, call
analyze directly. recordings.
1. Social Media Data – Twitter, Facebook, LinkedIn generate vast amounts of
user-generated content.
2. E-commerce Data – Websites like Amazon collect user preferences, purchase history.
3. IoT & Sensor Data – Smart devices and sensors collect data from machines,
wearables, smart cities.
4. Financial Data – Stock markets, credit card transactions, blockchain.
5. Healthcare Data – Electronic health records, medical imaging, patient histories.
Machine Learning & AI Helps find patterns, make predictions from big
data.
Cloud Computing (AWS, Google Provides scalable storage and processing power.
Cloud, Azure)