Data Analytics with Python Lecture 1: Introduction to Data Analytics

Objective of the Course

• The principle focus is on introducing conceptual understanding using simple and practical examples.
• Aims to make users comfortable with analytics in their career and life.
• Focuses on working with real data and choosing the right methodology.

Objective of the Course (Continued)

• The danger in using quantitative methods lies in the lack of fundamental understanding.
• Understanding includes:
  • Why to use a particular technique or procedure
  • How to use it correctly
  • How to correctly interpret the results

Learning Objectives

1. Define data and its importance.
2. Define data analytics and its types.
3. Explain why analytics is important in today’s business environment.
4. Explain how statistics, analytics, and data science are interrelated.
5. Why Python?
6. Explain the four different levels of Data:
   • Nominal
   • Ordinal
   • Interval
   • Ratio

Define Data and Its Importance

• Variable, Measurement and Data
• What is generating so much data?
• How does data add value to the business?
• Why is data important?

Variable, Measurement and Data

• Variables – A characteristic of any entity being studied that can take on different values.
• Measurements – A standard process used to assign numbers to attributes or characteristics of a variable.
• Data – Recorded measurements.

What is Generating So Much Data?

• Data can be generated by:
  • Humans
  • Machines
  • Human-machine combinations
• Generated wherever information is created and stored in structured or unstructured formats.

How Data Adds Value to Business

• Data warehouse development
• Development of Data Products
• Algorithm solutions in production, marketing, and sales (e.g., Recommendation Engines)
• Discovery of Data Insight
• Quantitative data analysis to help steer strategic business decisions

Data Products

• Recommendation Engines
  • Collaborative Filtering (recommending products based on similar customers)
  • Content Filtering (recommending products based on similar items)

Why Data is Important

• Helps in making better decisions.
• Helps in solving problems by finding the reasons for underperformance.
• Helps in evaluating performance.
• Helps in improving processes.
• Helps in understanding consumers and the market.

Define Data Analytics and Its Types

• Define data analytics
• Why analytics is important?
• Data analysis
• Data analytics vs. Data analysis
• Types of Data analytics

Define Data Analytics

• Analytics is defined as “the scientific process of transforming data into insights for making better decisions.”
• Analytics involves using data, information technology, statistical analysis, quantitative methods, and mathematical or computer-based models to help managers gain improved insight about their business operations and make better, fact-based decisions (James Evans).
• Analysis = Analytics?

Why Analytics is Important

• Opportunity abounds for the use of analytics and big data, such as:
  1. Determining credit risk
  2. Developing new medicines
  3. Finding more efficient ways to deliver products and services
  4. Preventing fraud
  5. Uncovering cyber threats
  6. Retaining the most valuable customers

Data Analysis

• Data analysis is the process of examining, transforming, and arranging raw data to generate useful information.
• Allows for the evaluation of data through analytical and logical reasoning to lead to some outcome or conclusion.
• It is a multi-faceted process with many steps, approaches, and diverse techniques.

Data Analytics vs. Data Analysis

• Analysis:
  • Past - Explains how and why something happened
  • Qualitative and Quantitative (Intuition + analysis)
  • Explains how and why the story ends the way it did.
  • Data + how the sale decreased last summer.
• Analytics:
  • Future - Explore potential future events
  • Qualitative and Quantitative (Formulas + algorithms)

Classification of Data Analytics

• Based on the phase of workflow, there are four major types:
  • Descriptive analytics
  • Diagnostic analytics
  • Predictive analytics
  • Prescriptive analytics

Descriptive Analytics

• Conventional form of Business Intelligence and data analysis.
• Seeks to provide a “summary view” of facts and figures, informing or preparing data for further analysis.
• Summarizes raw data into an easily understood format, describing events that have occurred in the past.
• Examples:
  • Data Queries
  • Reports
  • Descriptive Statistics
  • Data Visualization
  • Data dashboards

Diagnostic Analytics

• Examines data to answer the question “Why did it happen?”
• Aids in digging deeper into an issue to find the source of a problem.
• Tools for descriptive and diagnostic analytics often run parallel.
• Examples:
  1. Data Discovery
  2. Data Mining
  3. Correlations

Predictive Analytics

• Helps to forecast trends based on current events.
• Predicts the probability of future events or estimates when they will happen.
• Analyzes many co-dependent variables to predict trends. *Example:
  1. Linear regression
  2. Time series analysis and forecasting
  3. Data mining

Prescriptive Analytics

• Indicates the best course of action and what decisions to make to optimize outcomes.
• Aims to enable:
  1. Quality improvements
  2. Service enhancements
  3. Cost reductions
  4. Increasing productivity
     *Examples
  • Optimization Model
  • Simulation
  • Decision Analysis

Explain Why Analytics Is Important

• Demand for Data Analytics
• Element of data Analytics
• Data scientists are earning more than CAs and engineers.
• India will face a demand-supply gap of 200,000 analytics professionals over the next three years.
• Even in the US, only 40 out of 100 positions for analytics professionals can be filled.

Element of Data Analytics

• Business Intelligence/Information Systems
• Statistics
• Data Mining
• Simulation and Risk Modeling
• Visualization
• Optimization
• "What If?"

Data Analyst and Data Scientist

• The requisite skill set
• Difference between Data analyst and Data Scientist

The Requisite Skill Set

• Technology/Hacking Skill
• Business and strategy acumen
• Mathematic Expertise
• Data Science

Difference Between Data Analyst and Data Scientist

• Analyst
  • Business Administration
  • Domain-specific responsibility (e.g., marketing analyst, financial analyst)
  • Data exploration, analysis, and insight
• Data Scientist
  • Advance algorithms and machine learning
  • Data product engineering

Why Python?

Features

• Simple and easy to learn
• Freeware and Open source
• Interpreted
• Dynamically Typed
• Extensible
• Embedded
• Extensive library

Usability

• Desktop and web applications
• Database applications
• Networking applications
• Data analysis (Data Science)
• Machine learning
• IoT and AI applications
• Games

Companies Using Python

• Google
• NASA
• Yahoo
• eBay

Why Jupyter Notebook?

*Why?
* Client – Server Application
* Edit code on web browser
* Easy in documentation
* Easy in demonstration
* User- friendly Interface

Explain the Four Different Levels of Data

• Types of Variables
• Levels of Data Measurement
• Compare the four different levels of Data:
  • Nominal
  • Ordinal
  • Interval
  • Ratio
• Usage Potential of Various Levels of Data
• Data Level, Operations, and Statistical Methods

Types of Variables

Data

• Categorical
  • Examples:
    • Marital Status
    • Political Party
    • Eye Color (Defined categories)
• Numerical
  • Discrete
    • Examples:
      • Number of Children
      • Defects per hour (Counted items)
  • Continuous
    • Examples:
      • Weight
      • Voltage (Measured characteristics)

Levels of Data Measurement

• Nominal — Lowest level of measurement
• Ordinal
• Interval
• Ratio — Highest level of measurement

Nominal

• Classifies data into distinct categories where no ranking is implied.
• Example: Gender, Marital Status

Ordinal Scale

• Classifies data into distinct categories in which ranking is implied.
• Example:
  • Product satisfaction (Satisfied, Neutral, Unsatisfied)
  • Faculty rank (Professor, Associate Professor, Assistant Professor)
  • Student Grades (A, B, C, D, F)

Interval Scale

• An ordered scale where the difference between measurements is a meaningful quantity, but there is no true zero point.
• Example: Temperature in Fahrenheit and Celsius, Year

Ratio Scale

• An ordered scale where the difference between measurements is a meaningful quantity, and the measurements have a true zero point.
• Example: Weight, Age, Salary

Usage Potential of Various Levels of Data

• Ratio (highest)
• Interval
• Ordinal
• Nominal (lowest)

Impact of Choice of Measurement Scale