Multivariate Data Analysis Chapter 9 Cluster Analysis Section

Multivariate Data Analysis Chapter 9 - Cluster Analysis Section 3: Independence Techniques

Chapter 9 l What Is Cluster Analysis (Q analysis)? l l l Define groups of homogeneous objects (i. e. , individuals, firms, products, or behaviors) Maximize the homogeneity of objects within the clusters while also maximize the heterogeneity between clusters Segmentation and target marketing Compare with Factor Analysis How Does Cluster Analysis Work? l l l Measuring Similarity (Euclidean distance) Forming Clusters (hierarchical procedure vs. agglomerative method) Determining the Number of Clusters in the Final Solution (entropy group)

Cluster Analysis Decision Process l Stage One: Objectives of Cluster Analysis l l Taxonomy description Data simplification Relationship identification Selection of Clustering Variables l l Characterize the objects being clustered Relate specifically to the objectives of the cluster analysis

Cluster Analysis Decision Process (Cont. ) l Stage 2: Research Design in Cluster Analysis l l Detecting Outliers Similarity Measures (Interobject similarity) l Correlational Measures l Distance Measures § § § Association Measures Standardizing the Data l Standardizing By Variables (normalized distance function) l Standardizing By Observation (within-case vs. rowcentering standarlization) l l Comparison to Correlational Measures Types of Distance Measures (Euclidean distance) Impact of Unstandardized Data Values (Mahalonobis Distance, D 2)

Cluster Analysis Decision Process (Cont. ) l Stage 3: Assumptions in Cluster Analysis l l Representativeness of the Sample Impact of Multicollinearity

Cluster Analysis Decision Process (Cont. ) l Stage 4: Deriving Clusters and Assessing Overall Fit l Clustering Algorithms l Hierarchical Cluster Procedures § § § l Nonhierarchical Clustering Procedures § § l § l l Sequential Threshold Parallel Threshold Optimization Selecting Seed Points Should Hierarchical or Nonhierarchical Methods Be Used? § l Single Linkage Complete Linkage Average Linkage Ward's Method Centroid Method Pros and Cons of Hierarchical Methods Emergence of Nonhierarchical Methods A Combination of Both Methods How Many Clusters Should Be Formed? Should the Cluster Analysis Be Respecified

Cluster Analysis Decision Process (Cont. ) l l Stage 5: Interpretation of the Clusters Stage 6: Validation and Profiling of the Clusters l Validating the Cluster Solution l l l Criterion or predictive validity Profiling the Cluster Solution Summary of the Decision Process

An Illustrative Example l Stage 1: Objectives of the Cluster Analysis l l l Stage 2: Research Design of the Cluster Analysis l l l Segment objects (customers) into groups with similar perceptions of HATCO can then formulate strategies with different appeals for the separate groups. Identify any outliers Similarity measure (multicollinearity: D 2) Stage 3: Assumptions in Cluster Analysis

An Illustrative Example (Cont. ) l Stage 4: Deriving Clusters and Assessing Overall Fit l l l Stage 5: Interpretation of the Clusters l l Step 1: Hierarchical Cluster Analysis Step 2: Nonhierarchical Cluster Analysis Two-cluster solution Four-cluster solution Stage 6: Validation and Profiling of the Clusters Managerial view