1. Multivariate Data Analysis Chapter 5 – Discrimination Analysis and Logistic Regression

2. Chapter 5 What Are Discrimination Analysis and Logistic Regression? Analogy with Regression and MANOVA Hypothetical Example of Discriminant Analysis  A Two Group Discriminant Analysis: Purchasers Versus Nonpurchasers  A Geometric Representation of the Two Group Discriminant Function  A Three-group Example of Discriminant Analysis: Switching Intentions

3. The Decision Process for Discriminant Analysis Stage 1: Objectives of Discriminant Analysis Stage 2: Research Design for Discriminant Analysis  Selection of Dependent and Independent Variables  Sample Size  Division of the Sample Stage 3: Assumptions of Discriminant Analysis

4. The Decision Process for Discriminant Analysis (Cont.) Stage 4: Estimation of the Discriminant Model and Assessing Overall Fit  Computation Method  Statistical Significance  Assessing Overall Fit Calculating Discriminant Z Score Evaluating Group Differences Assessing Group Membership Prediction Accuracy  Casewise Diagnostics Misclassification of Individual Cases

5. The Decision Process for Discriminant Analysis (Cont.) Stage 5: Interpretation of the Results  Discriminant Weights  Discriminant Loadings  Partial F Values  Interpretation of Two or More Functions Rotation of the Discriminant Functions Potency Index Graphical Display of Discriminant Loadings  Which Interpretive Method to Use?

6. The Decision Process for Discriminant Analysis (Cont.) Stage 6: Validation of the Results  Split-Sample or Cross-Validation Procedures  Profiling Group Differences

7. Logistic Regression Regression with a Binary Dependent Variable  Representation of the Binary Dependent Variable  Estimating the Logistic Regression Model  Interpreting the Coefficients Assessing the Goodness-of-Fit of the Estimated Model Testing for Significance of the Coefficients Other Similarities To Multiple Regression