- ホーム
- > 洋書
- > 英文書
- > Business / Economics
Full Description
Crunch Big Data to optimize marketing and more!Overwhelmed by all the Big Data now available to you? Not sure what questions to ask or how to ask them? Using Microsoft Excel and proven decision analytics techniques, you can distill all that data into manageable sets-and use them to optimize a wide variety of business and investment decisions. In Decision Analytics: Microsoft Excel, best selling statistics expert and consultant Conrad Carlberg will show you how-hands-on and step-by-step.Carlberg guides you through using decision analytics to segment customers (or anything else) into sensible and actionable groups and clusters. Next, you'll learn practical ways to optimize a wide spectrum of decisions in business and beyond-from pricing to cross-selling, hiring to investments-even facial recognition software uses the techniques discussed in this book!Through realistic examples, Carlberg helps you understand the techniques and assumptions that underlie decision analytics and use simple Excel charts to intuitively grasp the results. With this foundation in place, you can perform your own analyses in Excel and work with results produced by advanced stats packages such as SAS and SPSS.This book comes with an extensive collection of downloadable Excel workbooks you can easily adapt to your own unique requirements, plus VBA code to streamline several of its most complex techniques.Classify data according to existing categories or naturally occurring clusters of predictor variables Cut massive numbers of variables and records down to size, so you can get the answers you really need Utilize cluster analysis to find patterns of similarity for market research and many other applications Learn how multiple discriminant analysis helps you classify cases Use MANOVA to decide whether groups differ on multivariate centroids Use principal components to explore data, find patterns, and identify latent factorsRegister your book for access to all sample workbooks, updates, and corrections as they become available at quepublishing.com/title/9780789751683.
Contents
Introduction 1What's in the Book 1Why Use Excel? 31 Components of Decision Analytics 5Classifying According to Existing Categories 5Using a Two-Step Approach 6Multiple Regression and Decision Analytics 6Access to a Reference Sample 8Multivariate Analysis of Variance 9Discriminant Function Analysis 10Logistic Regression 12Classifying According to Naturally Occurring Clusters 13Principal Components Analysis 13Cluster Analysis 14Some Terminology Problems 16The Design Sets the Terms 17Causation Versus Prediction 18Why the Terms Matter 182 Logistic Regression 21The Rationale for Logistic Regression 22The Scaling Problem 24About Underlying Assumptions 25Equal Spread 25Equal Variances with Dichotomies 27Equal Spread and the Range 28The Distribution of the Residuals 29Calculating the Residuals 30The Residuals of a Dichotomy 30Using Logistic Regression 31Using Odds Rather Than Probabilities 32Using Log Odds 33Using Maximum Likelihood Instead of Least Squares 34Maximizing the Log Likelihood 35Setting Up the Data 35Setting Up the Logistic Regression Equation 36Getting the Odds 38Getting the Probabilities 39Calculating the Log Likelihood 40Finding and Installing Solver 41Running Solver 41The Rationale for Log Likelihood 43The Probability of a Correct Classification 44Using the Log Likelihood 45The Statistical Significance of the Log Likelihood 48Setting Up the Reduced Model 50Setting Up the Full Model 513 Univariate Analysis of Variance (ANOVA) 53The Logic of ANOVA 54Using Variance 54Partitioning Variance 55Expected Values of Variances (Within Groups) 56Expected Values of Variances (Between Groups) 58The F-Ratio 61The Noncentral F Distribution 64Single Factor ANOVA 66Adopting an Error Rate 66Computing the Statistics 67Deriving the Standard Error of the Mean 70Using the Data Analysis Add-In 72Installing the Data Analysis Add-In 73Using the ANOVA: Single Factor Tool 73Understanding the ANOVA Output 75Using the Descriptive Statistics 75Using the Inferential Statistics 76The Regression Approach 79Using Effect Coding 80The LINEST() Formula 82The LINEST() Results 83LINEST() Inferential Statistics 854 Multivariate Analysis of Variance (MANOVA) 89The Rationale for MANOVA 89Correlated Variables 90Correlated Variables in ANOVA 91Visualizing Multivariate ANOVA 92Univariate ANOVA Results 93Multivariate ANOVA Results 93Means and Centroids 95From ANOVA to MANOVA 96Using SSCP Instead of SS 98Getting the Among and the Within SSCP Matrices 102Sums of Squares and SSCP Matrices 104Getting to a Multivariate F-Ratio 105Wilks' Lambda and the F-Ratio 107Converting Wilks' Lambda to an F Value 108Running a MANOVA in Excel 110Laying Out the Data 110Running the MANOVA Code 111Descriptive Statistics 112Equality of the Dispersion Matrices 113The Univariate and Multivariate F-Tests 115After the Multivariate Test 1165 Discriminant Function Analysis: The Basics 119Treating a Category as a Number 120The Rationale for Discriminant Analysis 122Multiple Regression and Discriminant Analysis 122Adjusting Your Viewpoint 123Discriminant Analysis and Multiple Regression 125Regression, Discriminant Analysis, and Canonical Correlation 125Coding and Multiple Regression 127The Discriminant Function and the Regression Equation 129From Discriminant Weights to Regression Coefficients 130Eigenstructures from Regression and Discriminant Analysis 133Structure Coefficients Can Mislead 136Wrapping It Up 1376 Discriminant Function Analysis: Further Issues 139Using the Discriminant Workbook 139Opening the Discriminant Workbook 140Using the Discriminant Dialog Box 141Why Run a Discriminant Analysis on Irises? 144Evaluating the Original Measures 144Discriminant Analysis and Investment 145Benchmarking with R 147Downloading R 147Arranging the Data File 148Running the Analysis 149The Results of the Discrim Add-In 152The Discriminant Results 153Interpreting the Structure Coefficients 155Eigenstructures and Coefficients 156Other Uses for the Coefficients 159Classifying the Cases 162Distance from the Centroids 163Correcting for the Means 164Adjusting for the Variance-Covariance Matrix 167Assigning a Classification 169Creating the Classification Table 170Training Samples: The Classification Is Known Beforehand 1717 Principal Components Analysis 173Establishing a Conceptual Framework for Principal Components Analysis 174Principal Components and Tests 174PCA's Ground Rules 175Correlation and Oblique Factor Rotation 176Using the Principal Components Add-In 177The Correlation Matrix 179The Inverse of the R Matrix 179The Sphericity Test 182Counting Eigenvalues, Calculating Coefficients and Understanding Communalities 183How Many Components? 184Factor Score Coefficients 186Communalities 186Relationships Between the Individual Results 187Using the Eigenvalues and Eigenvectors 187Eigenvalues, Eigenvectors, and Loadings 188Eigenvalues, Eigenvectors, and Factor Coefficients 190Getting the Eigenvalues Directly from the Factor Scores 191Getting the Eigenvalues and Eigenvectors 192Iteration and Exhaustion 193Rotating Factors to a Meaningful Solution 196Identifying the Factors 197The Varimax Rotation 200Classification Examples 202State Crime Rates 202Physical Measurements of Aphids 2068 Cluster Analysis: The Basics 209Cluster Analysis, Discriminant Analysis, and Logistic Regression 209Euclidean Distance 211Mahalanobis' D2 and Cluster Analysis 214Finding Clusters: The Single Linkage Method 215The Self-Selecting Nature of Cluster Analysis 220Finding Clusters: The Complete Linkage Method 223Complete Linkage: An Example 224Other Linkage Methods 227Finding Clusters: The K-means Method 228Characteristics of K-means Analysis 228A K-means Example 229Benchmarking K-means with R 2339 Cluster Analysis: Further Issues 235Using the K-means Workbook 235Deciding on the Number of Clusters 237The Cluster Members Worksheet 239The Cluster Centroids Worksheet 241The Cluster Variances Worksheet 242The F-Ratios Worksheet 244Reporting Process Statistics 247Cluster Analysis Using Principal Components 248Principal Components Revisited 249Clustering Wines 253Cross-Validating the Results 256Index 259
