Applying predictive analytics : finding value in data

Preface
Contents
About the Authors
Chapter 1: Introduction to Predictive Analytics
	1.1 Predictive Analytics in Action
	1.2 Analytics Landscape
		1.2.1 Big Data
	1.3 Analytics
		1.3.1 Predictive Analytics
	1.4 Regression Analysis
	1.5 Machine Learning Techniques
	1.6 Predictive Analytics Model
	1.7 Opportunities in Analytics
	1.8 Introduction to the Automobile Insurance Claim Fraud Example
	1.9 Chapter Summary
	References
Chapter 2: Know Your Data: Data Preparation
	2.1 Classification of Data
		2.1.1 Qualitative Versus Quantitative
		2.1.2 Scales of Measurement
	2.2 Data Preparation Methods
		2.2.1 Inconsistent Formats
		2.2.2 Missing Data
		2.2.3 Outliers
		2.2.4 Other Data Cleansing Considerations
	2.3 Data Sets and Data Partitioning
	2.4 SAS Enterprise Miner™ Model Components
		2.4.1 Step 1. Create Three of the Model Components
		2.4.2 Step 2. Import an Excel File and Save as a SAS File
		2.4.3 Step 3. Create the Data Source
		2.4.4 Step 4. Partition the Data Source
		2.4.5 Step 5. Data Exploration
		2.4.6 Step 6. Missing Data
		2.4.7 Step 7. Handling Outliers
		2.4.8 Step 8. Categorical Variables with Too Many Levels
	2.5 Chapter Summary
	Reference
Chapter 3: What Do Descriptive Statistics Tell Us
	3.1 Descriptive Analytics
	3.2 The Role of the Mean, Median, and Mode
	3.3 Variance and Distribution
	3.4 The Shape of the Distribution
		3.4.1 Skewness
		3.4.2 Kurtosis
	3.5 Covariance and Correlation
	3.6 Variable Reduction
		3.6.1 Variable Clustering
		3.6.2 Principal Component Analysis
	3.7 Hypothesis Testing
	3.8 Analysis of Variance (ANOVA)
	3.9 Chi-Square
	3.10 Fit Statistics
	3.11 Stochastic Models
	3.12 Chapter Summary
	References
Chapter 4: Predictive Models Using Regression
	4.1 Regression
		4.1.1 Classical Assumptions
	4.2 Ordinary Least Squares
	4.3 Simple Linear Regression
		4.3.1 Determining Relationship Between Two Variables
		4.3.2 Line of Best Fit and Simple Linear Regression Equation
	4.4 Multiple Linear Regression
		4.4.1 Metrics to Evaluate the Strength of the Regression Line
		4.4.2 Best-Fit Model
		4.4.3 Selection of Variables in Regression
	4.5 Principal Component Regression
		4.5.1 Principal Component Analysis Revisited
		4.5.2 Principal Component Regression
	4.6 Partial Least Squares
	4.7 Logistic Regression
		4.7.1 Binary Logistic Regression
		4.7.2 Examination of Coefficients
		4.7.3 Multinomial Logistic Regression
		4.7.4 Ordinal Logistic Regression
	4.8 Implementation of Regression in SAS Enterprise Miner™
		4.8.1 Regression Node Train Properties: Class Targets
		4.8.2 Regression Node Train Properties: Model Options
		4.8.3 Regression Node Train Properties: Model Selection
	4.9 Implementation of Two-Factor Interaction and Polynomial Terms
		4.9.1 Regression Node Train Properties: Equation
	4.10 DMINE Regression in SAS Enterprise Miner™
		4.10.1 DMINE Properties
		4.10.2 DMINE Results
	4.11 Partial Least Squares Regression in SAS Enterprise Miner™
		4.11.1 Partial Least Squares Properties
		4.11.2 Partial Least Squares Results
	4.12 Least Angle Regression in SAS Enterprise Miner™
		4.12.1 Least Angle Regression Properties
		4.12.2 Least Angle Regression Results
	4.13 Other Forms of Regression
	4.14 Chapter Summary
	References
Chapter 5: The Second of the Big 3: Decision Trees
	5.1 What Is a Decision Tree?
	5.2 Creating a Decision Tree
	5.3 Classification and Regression Trees (CART)
	5.4 Data Partitions and Decision Trees
	5.5 Creating a Decision Tree Using SAS Enterprise Miner™
		5.5.1 Overfitting
	5.6 Creating an Interactive Decision Tree Using SAS Enterprise Miner™
	5.7 Creating a Maximal Decision Tree Using SAS Enterprise Miner™
	5.8 Chapter Summary
	References
Chapter 6: The Third of the Big 3: Neural Networks
	6.1 What Is a Neural Network?
	6.2 History of Neural Networks
	6.3 Components of a Neural Network
	6.4 Neural Network Architectures
	6.5 Training a Neural Network
	6.6 Radial Basis Function Neural Networks
	6.7 Creating a Neural Network Sing SAS Enterprise Miner™
	6.8 Using SAS Enterprise Miner™ to Automatically Generate a Neural Network
	6.9 Explaining a Neural Network
	6.10 Chapter Summary
	References
Chapter 7: Model Comparisons and Scoring
	7.1 Beyond the Big 3
	7.2 Gradient Boosting
	7.3 Ensemble Models
	7.4 Random Forests
	7.5 Memory-Based Reasoning
	7.6 Two-Stage Model
	7.7 Comparing Predictive Models
		7.7.1 Evaluating Fit Statistics: Which Model Do We Use?
	7.8 Using Historical Data to Predict the Future: Scoring
		7.8.1 Analyzing and Reporting Results
		7.8.2 Save Data Node
		7.8.3 Reporter Node
	7.9 The Importance of Predictive Analytics
		7.9.1 What Should We Expect for Predictive Analytics in the Future?
	7.10 Chapter Summary
	References
Chapter 8: Finding Associations in Data Through Cluster Analysis
	8.1 Applications and Uses of Cluster Analysis
	8.2 Types of Clustering Techniques
	8.3 Hierarchical Clustering
		8.3.1 Agglomerative Clustering
		8.3.2 Divisive Clustering
		8.3.3 Agglomerative Versus Divisive Clustering
	8.4 Non-hierarchical Clustering
		8.4.1 K-Means Clustering
		8.4.2 Initial Centroid Selection
		8.4.3 Determining the Number of Clusters
		8.4.4 Evaluating Your Clusters
	8.5 Hierarchical Versus Non-hierarchical
	8.6 Cluster Analysis Using SAS Enterprise Miner™
		8.6.1 Cluster Node
		8.6.2 Additional Key Properties of the Cluster Node
	8.7 Applying Cluster Analysis to the Insurance Claim Fraud Data Set
	8.8 Chapter Summary
	References
Chapter 9: Text Analytics: Using Qualitative Data to Support Quantitative Results
	9.1 What Is Text Analytics?
	9.2 Information Retrieval
	9.3 Text Parsing
	9.4 Zipf’s Law
	9.5 Text Filter
	9.6 Text Cluster
	9.7 Text Topic
	9.8 Text Rule Builder
	9.9 Text Profile
	9.10 Chapter Summary
	References
Appendix A: Data Dictionary for the Automobile Insurance Claim Fraud Data Example
Appendix B: Can You Predict the Money Laundering Cases?
	Introduction
	Business Problem
	Analyze Data
	Development and Optimization of a Best-Fit Model
	Final Report
References
Index