Foundational Concepts of Business Analytics

Sep 23, 2024

Take quiz

Review flashcards

Business Analytics Lecture Notes

Introduction to Business Analytics

  • Definition: Process of transforming data into insights to support, improve, or automate business decisions.
  • Data Types and Sources: Includes internal (e.g., sensors, barcodes, websites) and external data sources (e.g., government surveys, social media).
  • Techniques and Software: Various techniques and software packages available for analysis.

Wisdom Hierarchy (DIKW Pyramid)

  • Data: Numbers or text without context.
  • Information: Provides meaning from data, often combining multiple data points.
  • Knowledge: Contextualizes information, making it applicable to situations.
  • Wisdom: Applies knowledge to make decisions; data becomes useful for action.

Data Sources and Growth

  • Internal data: Collected by businesses, stored within their servers.
  • External data: Public domain (e.g., surveys, social media) and paid sources (e.g., stock market, weather data).
  • Growth: Data generation and collection are growing exponentially due to increased computing power and cheaper storage.

Business Analytics Terms and Applications

  • Synonymous terms: Data Science, Business Intelligence, Big Data, Data Mining, Knowledge Discovery, Machine Learning.
  • Interdisciplinary field: Combines math, statistics, and computer science principles.
  • Applications: Retail analytics, marketing campaigns, supply chain optimization, staffing, pricing, sports analytics.

History and Evolution of Analytics

  • Early use: From WWII decoding to weather forecasting (1950s) and credit risk modeling (1958).
  • Transition: From historical projects to real-time analytics (e.g., real-time credit card fraud detection in 1992).
  • Tech companies: Google and Amazon as examples of analytically centric companies.

Key Areas for Business Use

  • Competition: Increase revenue.
  • Efficiency: Reduce costs.
  • Customer Satisfaction: Improve experience and loyalty.

Case Study: Grocery Stores

  • Loyalty cards for customer identification and targeted promotions.
  • Store layout and shelf arrangement for maximizing spending.
  • Pricing strategies like loss leaders and markdowns to prevent losses.

Role of Business Analyst

  • Career appeal: High demand, interesting problem-solving, uncovering unknown truths.
  • Roles: Interpreter (descriptive analytics), Oracle (predictive analytics), Console (prescriptive analytics).

Key Skills for Success

  • Hard Skills: Analytical tools, coding knowledge.
  • Soft Skills: Equally or more important than hard skills.

Analytical Tools

  • Coding-required software: SAS, R, Python, SQL.
  • GUI-based tools: Tableau, Alteryx, RapidMiner.

Project Goals and Outcomes

  • Information delivery: Reports, dashboards for stakeholders.
  • Analytical products: Automated algorithms for consumer experience and business efficiency.

Analytical Process Steps

  1. Define project goals.
  2. Collect and inventory data sources.
  3. Understand data through observation and questioning.
  4. Explore and prepare data (data cleaning, feature engineering).
  5. Build models and put them into production.

Importance of Data Visualization

  • Visualization aids throughout the analytical process.

Foundations of Business Analytics

  • Selecting Variables: Identify and keep only relevant fields.
  • Filtering Data: Condition subsets to omit certain rows.
  • Sorting Data: Arrange data for insights.

Formulas and Data Preparation

  • Single-row formulas: Numeric calculations, string transformations.
  • Multi-row formulas: Running totals, lag/lead values, window functions.

Data Combining Techniques

  • Unions: Combine tables vertically.
  • Joins: Merge tables using keys (inner join, left outer join).

Aggregation and Summarization

  • Use aggregation to answer basic data questions.

Crosstabs and Transposing

  • Crosstabs: Compare measures across dimensions.
  • Transposing: Convert wide format data to narrow format.

Statistical Concepts

  • Contingency Tables: Display multivariate frequency distributions.
  • Distributions: Analyze numeric variable distributions.
  • Variation: Understand spread using range, variance, standard deviation.
  • Normal Distributions: Symmetric bell-shaped curve properties.
  • Kurtosis and Skewness: Tail size and asymmetrical distributions.

Sampling and Bivariate Data Analysis

  • Sampling Methods: Random, stratified, cluster sampling.
  • Bivariate Data: Analyze relationships with scatterplots.

Uncertainty and Entropy

  • Measure unpredictability and resolve uncertainty through data analysis.

Writing Analytical Reports

  • Structure: Introduction, data analysis, results, and conclusion.

Automation in Analytics

  • Macros: Replicate steps efficiently.
  • Stored Procedures: Automate scheduled tasks.

Regression Analysis

  • Simple Linear Regression: Predict one variable based on another.
  • Logistic Regression: Model dichotomous outcomes.

Statistical Errors and Hypothesis Testing

  • Type 1 & 2 Errors: False positives and negatives.
  • Hypothesis Testing Steps: State hypothesis, locate critical region, compute test statistic, draw conclusions.

Correlation Analysis

  • Measure and interpret the degree of association between variables.

Probability and Variables

  • Probability Rules: Complements, addition rule, independence.
  • Variable Classification: Qualitative and quantitative types.

Coding and Data Preparation

  • Proper coding ensures data consistency and clarity.

Summary of Key Statistical Measures

  • Central Tendency: Mean, median, mode.
  • Variation: Range, quartiles, variance, standard deviation.

These notes provide a comprehensive overview of the foundational concepts and applications of business analytics, essential for understanding and applying analytical techniques in various business contexts.