Business Analytics: Data Science for Business Problems

Description

This book focuses on three core knowledge requirements for effective and thorough data analysis for solving business problems. These are a foundational understanding of: 1. statistical, econometric, and machine learning techniques; 2. data handling capabilities; 3. at least one programming language. Practical in orientation, the volume offers illustrative case studies throughout and examples using Python in the context of Jupyter notebooks. Covered topics include demand measurement and forecasting, predictive modeling, pricing analytics, customer satisfaction assessment, market and advertising research, and new product development and research. This volume will be useful to business data analysts, data scientists, and market research professionals, as well as aspiring practitioners in business data analytics. It can also be used in colleges and universities offering courses and certifications in business data analytics, data science, and market research. Preface The Book's Focus The Target Audience The Book's Competitive Comparison The Book's Structure Acknowledgments Contents List of Figures List of Tables Part I Beginning Analytics 1 Introduction to Business Data Analytics: Setting the Stage 1.1 Types of Business Problems 1.2 The Role of Information in Business Decision Making 1.3 Uncertainty vs. Risk 1.4 The Data-Information Nexus 1.4.1 Data and Information Confusion 1.4.2 The Data Component 1.4.3 The Extractor Component 1.4.3.1 Text Data 1.4.3.2 Numeric Data 1.4.3.3 Data: A Combined View 1.4.4 The Information Component 1.5 Analytics Requirements 1.5.1 Theoretical Framework 1.5.2 Data Handling 1.5.3 Programming Literacy 1.5.4 Component Interconnections 2 Data Sources, Organization, and Structures 2.1 Data Dimensions: A Taxonomy for Defining Data 2.1.1 Taxonomy Component #1: Source 2.1.2 Taxonomy Component #2: Domain 2.1.3 Taxonomy Component #3: Levels 2.1.4 Taxonomy Component #4: Continuity 2.1.5 Taxonomy Component #5: Measurement Scale 2.2 Data Organization 2.2.1 External Database Structures 2.2.2 Internal Database Structures 2.3 Data Dictionary 3 Basic Data Handling 3.1 Case Studies 3.1.1 Case Study 1: Customer Transactions Data 3.1.2 Case Study 2: Measures of Order Fulfillment 3.2 Importing Your Data 3.2.1 Data Formats 3.2.2 Importing a CSV Text File into Pandas 3.2.3 Importing Large Files in Chunks 3.2.4 Checking Your Imported Data 3.2.4.1 Check #1: Display the First Few Records 3.2.4.2 Check #2: Check the Shape of the DataFrame 3.2.4.3 Check #3: Check Column Names 3.2.4.4 Check #4: Check for Missing Values 3.2.4.5 Check #5: Check the Data Types 3.3 Merging or Joining DataFrames 3.4 Reshaping DataFrames 3.5 Sorting a DataFrame 3.6 Querying a DataFrame 3.6.1 Boolean Operators and Indicator Functions 3.6.2 Pandas Query Method 4 Data Visualization: The Basics 4.1 Background for Data Visualization 4.2 Gestalt Principles of Visual Design 4.3 Issues Complicating Data Visualization 4.3.1 Human Visual Limitations 4.3.2 Data Visualization Tools 4.3.3 Types of Visuals 4.3.4 What to Look for in a Graph 4.3.4.1 Feature #1: Distributions 4.3.4.2 Feature #2: Relationships 4.3.4.3 Feature #3: Patterns 4.3.4.4 Feature #4: Trends 4.3.4.5 Feature #5: Anomalies 4.4 Visualizing Spatial Data 4.4.1 Data Preparation 4.4.2 Visualizing Continuous Spatial Data 4.4.3 Visualizing Categorical Spatial Data 4.4.4 Visualizing Continuous and Categorical Spatial Data 4.5 Visualizing Temporal (Time Series) Data 4.5.1 Properties of Temporal (Time Series) Data 4.5.2 Visualizing Time Series Data 4.5.3 Times Series Complications 4.6 Faceted Plots 4.7 Appendix 4.7.1 Taylor Series Expansion for Growth Rates 5 Advanced Data Handling: Preprocessing Methods 5.1 Transformations 5.1.1 Linear Transformations 5.1.2 Nonlinear Transformations 5.1.3 A Family of Transformations 5.2 Encoding 5.2.1 Dummy or One-Hot Encoding 5.2.1.1 Pandas Dummy Encoding 5.2.1.2 sklearn Dummy Encoding 5.2.2 Patsy Encoding 5.2.3 Label Encoding 5.2.4 Binarizing Data 5.3 Dimension Reduction 5.4 Handling Missing Data 5.5 Appendix 5.5.1 Mean and Variance of Standardized Variable 5.5.2 Mean and Variance of Adjusted Standardized Variable 5.5.3 Unbiased Estimators of μ and σ2 Part II Intermediate Analytics 6 OLS Regression: The Basics 6.1 Basic OLS Concept 6.1.1 The Disturbance Term and the Residual 6.1.2 OLS Estimation 6.1.3 The Gauss-Markov Theorem 6.2 Analysis of Variance 6.3 Case Study 6.3.1 Basic OLS Regression 6.3.2 The Log-Log Model 6.3.3 Model Set-up 6.3.4 Estimation Summary 6.3.5 ANOVA for Basic Regression 6.3.6 Elasticities 6.4 Basic Multiple Regression 6.4.1 ANOVA for Multiple Regression 6.4.2 Alternative Measures of Fit: AIC and BIC 6.5 Case Study: Expanded Analysis 6.6 Model Portfolio 6.7 Predictive Analysis: Introduction 6.7.1 Predicting vs. Forecasting 6.7.2 Developing a Prediction 6.7.3 Simulation Tool for Prediction Application 7 Time Series Analysis 7.1 Time Series Basics 7.1.1 Time Series Definition 7.1.2 Time Series Concepts 7.2 Importing a Date/Time Variable 7.3 The Data Cube and Time Series Data 7.4 Handling Dates and Times in Python and Pandas 7.4.1 Datetimes vs. Periods 7.4.2 Aggregating Datetime Measures 7.4.3 Converting Time Periods in Pandas 7.4.4 Date-Time Mini-Language 7.5 Some Calendrical Calculations 7.6 Time Series Generation Process: AR(1) Model 7.7 Visualization for AR(1) Detection 7.8 Durbin-Watson Test Statistic 7.9 Lagged Dependent and Independent Variables 7.9.1 Lagged Independent Variable: ARDL(0, 1) 7.9.2 Lagged Dependent Variable: ARDL(1, 0) 7.9.3 Lagged Dependent and Independent Variables:ARDL(1, 1) 7.10 Further Exploration of Time Series Analysis 7.10.1 Step 1: Identification of a Model 7.10.1.1 AR(p) Model 7.10.1.2 MA(q) Model 7.10.1.3 ARMA(p, q) Model 7.10.1.4 ARIMA(p, d, q) Model 7.10.1.5 Digression: Time Series Stationarity—An Overview 7.10.2 Step 2: Estimation of the Model 7.10.3 Step 3: Validation of the Model 7.10.4 Step 4: Forecasting with the Model 7.11 Appendix 7.11.1 Backshift Operator 7.11.2 Useful Algebra Results 7.11.3 Mean and Variance of Yt 7.11.4 Demeaned Data 7.11.5 Time Trend Addition 8 Statistical Tables 8.1 Data Preprocessing 8.2 Categorical Data 8.3 Creating a Frequency Table 8.4 Hypothesis Testing: A First Step 8.5 Cross-tabs and Hypothesis Tests 8.5.1 Hypothesis Testing 8.5.2 Plotting a Frequency Table 8.6 Extending the Cross-tab 8.7 Pivot Tables 8.8 Appendix 8.8.1 Pearson Chi-Square Statistic Part III Advanced Analytics 9 Advanced Data Handling for Business Data Analytics 9.1 Supervised and Unsupervised Learning 9.2 Working with the Data Cube 9.3 The Data Cube and DataFrame Indexing 9.4 Sampling From a DataFrame 9.4.1 Simple Random Sampling (SRS) 9.4.2 Stratified Random Sampling 9.4.3 Cluster Random Sampling 9.5 Index Sorting of a DataFrame 9.6 Splitting a DataFrame: The Train-Test Splits 9.6.1 Model Tuning of Hyperparameters 9.6.2 Incorrect Use of Testing Data 9.6.3 Creating the Training/Testing Data Sets 9.6.3.1 Comment on Strategy 9.6.3.2 Handling Cross-Sectional Data 9.6.3.3 Handling Time Series Data 9.6.3.4 Handling Panel Data 9.6.4 Recombining the Data Sets 9.7 Appendix 9.7.1 Primer on Random Numbers 10 Advanced OLS for Business Data Analytics 10.1 Link Functions: An Introduction 10.2 Data Preprocessing 10.2.1 Data Standardization for Regression Analysis 10.2.2 One-Hot and Effects (or Sum) Encoding 10.3 Case Study Application 10.4 Heteroskedasticity Issues and Tests 10.4.1 Heteroskedasticity Problem 10.4.2 Heteroskedasticity Detection 10.4.3 Heteroskedasticity Remedy 10.5 Multicollinearity 10.5.1 Digression on Multicollinearity 10.5.2 Detection with VIF and the Condition Index 10.5.3 Principal Component Regression and High-Dimensional Data 10.6 Predictions and Scenario Analysis 10.6.1 Making Predictions 10.6.2 Scenario Analysis 10.6.3 Prediction Error Analysis (PEA) 10.6.3.1 LOOCV Approach 10.6.3.2 k-Fold Approach 10.6.3.3 Score Measures 10.6.3.4 Variations on Validation Methods 10.6.3.5 Complexity of Testing 10.6.3.6 Examples of k-Fold Split 10.7 Panel Data Models 11 Classification with Supervised Learning Methods 11.1 Case Study: Background 11.2 Logistic Regression 11.2.1 A Choice Interpretation 11.2.2 Properties of this Problem 11.2.3 A Model for the Binary Problem 11.2.4 Case Study: Train-Test Data Split 11.2.5 Case Study: Logit Model Training 11.2.6 Making and Assessing Predictions 11.2.7 Classification with a Logit Model 11.3 K-Nearest Neighbor (KNN) 11.3.1 Case Study: Predicting 11.4 Naive Bayes 11.4.1 Background: Bayes Theorem 11.4.2 A General Statement 11.4.3 The Naive Adjective: A Simplifying Assumption 11.4.4 Distribution Assumptions 11.4.5 Case Study: Naive Bayes Training 11.5 Decision Trees for Classification 11.5.1 Partitioning by Constants 11.5.2 Gini Index and Entropy 11.5.3 Case Study: Growing a Tree 11.5.4 Case Study: Predicting with a Tree 11.5.5 Random Forests 11.6 Support Vector Machines 11.6.1 Case Study: SVC Application 11.6.2 Case Study: Prediction 11.7 Classifier Accuracy Comparison 12 Grouping with Unsupervised Learning Methods 12.1 Training and Testing Data Sets 12.2 Hierarchical Clustering 12.2.1 Forms of Hierarchical Clustering 12.2.2 Agglomerative Algorithm Description 12.2.3 Metrics and Linkages 12.2.4 Preprocessing Data 12.2.5 Case Study Application 12.2.6 Examining More than One Solution 12.3 K-Means Clustering 12.3.1 Algorithm Description 12.3.2 Case Study Application 12.4 Mixture Model Clustering Bibliography Index