List of Figures

Chapter 2: Introduction to Regression Procedures

Figure 2.1: Regression for Weight and Height Data

Figure 2.2: Regression for Weight and Height Data

Figure 2.3: Default Multivariate Tests

Figure 2.4: Multivariate Tests with MSTAT=EXACT

Chapter 6: Introduction to Discriminant Procedures

Figure 6.1: Groups for Contrasting Univariate and Multivariate Analyses

Figure 6.2: Contrasting Univariate and Multivariate Analyses

Chapter 7: Introduction to Clustering Procedures

Figure 7.1: Data Containing Well-Separated, Compact Clusters: PROC CLUSTER with METHOD=SINGLE and PROC GPLOT

Figure 7.2: Data Containing Poorly Separated, Compact Clusters: Plot of True Clusters

Figure 7.3: Data Containing Poorly Separated, Compact Clusters: PROC FASTCLUS

Figure 7.4: Data Containing Poorly Separated, Compact Clusters: PROC CLUSTER with METHOD=WARD

Figure 7.5: Data Containing Poorly Separated, Compact Clusters: PROC CLUSTER with METHOD=AVERAGE

Figure 7.6: Data Containing Poorly Separated, Compact Clusters: PROC CLUSTER with METHOD=CENTROID

Figure 7.7: Data Containing Poorly Separated, Compact Clusters: PROC CLUSTER with METHOD=TWOSTAGE

Figure 7.8: Data Containing Poorly Separated, Compact Clusters: PROC CLUSTER with METHOD=SINGLE

Figure 7.9: Data Containing Generated Clusters of Unequal Size

Figure 7.10: Data Containing Compact Clusters of Unequal Size: PROC FASTCLUS

Figure 7.11: Data Containing Compact Clusters of Unequal Size: PROC CLUSTER with METHOD=WARD

Figure 7.12: Data Containing Compact Clusters of Unequal Size: PROC CLUSTER with METHOD=AVERAGE

Figure 7.13: Data Containing Compact Clusters of Unequal Size: PROC CLUSTER with METHOD=CENTROID

Figure 7.14: Data Containing Compact Clusters of Unequal Size: PROC CLUSTER with METHOD=TWOSTAGE

Figure 7.15: Data Containing Compact Clusters of Unequal Size: PROC CLUSTER with METHOD=SINGLE

Figure 7.16: Data Containing Parallel Elongated Clusters: PROC FASTCLUS

Figure 7.17: Data Containing Parallel Elongated Clusters: PROC CLUSTER with METHOD=AVERAGE

Figure 7.18: Data Containing Parallel Elongated Clusters: PROC CLUSTER with METHOD=TWOSTAGE

Figure 7.19: Data Containing Parallel Elongated Clusters: PROC ACECLUS

Figure 7.20: Data Containing Parallel Elongated Clusters After Transformation by PROC ACECLUS

Figure 7.21: Transformed Data Containing Parallel Elongated Clusters: PROC CLUSTER with METHOD=WARD

Figure 7.22: Data Containing Nonconvex Clusters: PROC FASTCLUS

Figure 7.23: Data Containing Nonconvex Clusters: PROC CLUSTER with METHOD=CENTROID

Figure 7.24: Data Containing Nonconvex Clusters: PROC CLUSTER with METHOD=TWOSTAGE

Chapter 13: Introduction to Structural Equation Modeling

Figure 13.1: Measurement Error Model for Corn Data

Figure 13.2: Spleen Data: Parameter Estimates for Overidentified Model

Figure 13.3: Spleen Data: Fit Statistics for Overidentified Model

Figure 13.4: Spleen Data: Parameter Estimated for Just Identified Model

Figure 13.5: Path Diagram: Spleen

Figure 13.6: Path Diagram: Spleen

Figure 13.7: Spleen Data: RAM Model

Figure 13.8: Spleen Data: RAM Model with Names for Latent Variables

Figure 13.9: Spleen Data: OUTRAM= Data Set with Final Parameter Estimates

Figure 13.10: Spleen Data: RAM Model with INRAM= Data Set

Figure 13.11: Path Diagram: Lord

Figure 13.12: Lord Data: Major Results for RAM Model, Hypothesis H4

Figure 13.13: Lord Data: Using LINEQS Statement for RAM Model, Hypothesis H4

Figure 13.14: Lord Data: Major Results for Hypothesis H3

Figure 13.15: Lord Data: Major Results for Hypothesis H2

Figure 13.16: Lord Data: Major Results for Hypothesis H1

Figure 13.17: Path Diagram: Career Aspiration Jreskog and Srbom

Figure 13.18: Career Aspiration Data: J&S Analysis 1

Figure 13.19: Career Aspiration Data: J&S Analysis 2

Figure 13.20: Path Diagram: Career Aspiration Loehlin

Figure 13.21: Career Aspiration Data: Loehlin Model 1

Figure 13.22: Career Aspiration Data: Loehlin Model 2

Figure 13.23: Career Aspiration Data: Loehlin Model 3

Figure 13.24: Career Aspiration Data: Loehlin Model 4

Figure 13.25: Career Aspiration Data: Loehlin Model 5

Figure 13.26: Career Aspiration Data: Loehlin Model 7

Figure 13.27: Career Aspiration Data: Loehlin Model 6

Figure 13.28: Career Aspiration Data: Model Comparisons

Chapter 14: Using the Output Delivery System

Figure 14.1: Partial Contents of the SAS Log: Result of the ODS TRACE ON Statement

Figure 14.2: The Results Window from the SAS Explorer

Chapter 15: Statistical Graphics Using ODS (Experimental)

Figure 15.1: Fit Diagnostics Panel

Figure 15.2: Residual Plot

Figure 15.3: Fit Plot

Figure 15.4: Contour Plot of Estimated Density

Figure 15.5: Surface Plot of Estimated Density

Figure 15.6: Current Folder (Right Bottom)

Figure 15.7: Disabling View of Results as Generated

Figure 15.8: Selecting an External Browser

Figure 15.9: Changing the Default External Browser

Figure 15.10: ODS Trace Record in SAS Log

Figure 15.11: HTML Output with Default Style

Figure 15.12: HTML Output with Journal Style

Figure 15.13: Requesting the Templates Window in the Command Line

Figure 15.14: Result of ODS PATH SHOW Statement

Figure 15.15: SAS Registry Editor

Figure 15.16: Selecting a Default Style for HTML Destination

Figure 15.17: Label Collision Avoidance

Chapter 16: The ACECLUS Procedure

Figure 16.1: Scatter Plot of Original Poverty Data: Birth Rate versus Death Rate

Figure 16.2: Means, Standard Deviations, and Covariance Matrix from the ACECLUS Procedure

Figure 16.3: Table of Iteration History from the ACECLUS Procedure

Figure 16.4: Approximate WithinCluster Covariance Estimates

Figure 16.5: Raw and Standardized Canonical Coefficients from the ACECLUS Procedure

Figure 16.6: Scatter Plot of Poverty Data, Identified by Cluster

Figure 16.7: Scatter Plot of Canonical Variables

Chapter 17: The ANOVA Procedure

Figure 17.1: Class Level Information

Figure 17.2: ANOVA Table

Figure 17.3: Tukeys Multiple Comparisons Procedure

Figure 17.4: Box Plot of Nitrogen Content for each Treatment (Experimental)

Figure 17.5: Class Level Information

Figure 17.6: Overall ANOVA Table for Yield

Figure 17.7: Tests of Effects for Yield

Figure 17.8: ANOVA Table for Worth

Figure 17.9: Means of Yield and Worth

Chapter 18: The BOXPLOT Procedure

Figure 18.1: Box Plot for Power Output Data

Figure 18.2: Box Plot with Insets

Figure 18.3: Skeletal Box-and-Whisker Plot

Figure 18.4: Box Plot: the NOTCHES Option

Figure 18.5: BOXSTYLE= SCHEMATIC

Figure 18.6: Box Plot with Discrete Group Variable

Figure 18.7: Box Plot with Continuous Group Variable

Figure 18.8: Insets Positioned Using Compass Points

Figure 18.9: Positioning Insets in the Margins

Figure 18.10: Inset Positioned Using Data Unit Coordinates

Figure 18.11: Inset Positioned Using Axis Percent Unit Coordinates

Figure 18.12: Box Plot Using a Block Variable

Figure 18.13: Compressed Box Plots

Figure 18.14: Box Plot with Clip Factor of 1.5

Figure 18.15: Box Plot Using Clipping Options

Chapter 19: The CALIS Procedure

Figure 19.1: Path Diagram of Stability and Alienation Example

Figure 19.2: Path Diagram of Second-Order Factor Analysis Model

Figure 19.3: Examples of RAM Nomography

Figure 19.4: Within-List and Between-List Covariances

Figure 19.5: Exogenous and Endogenous Variables

Chapter 20: The CANCORR Procedure

Figure 20.1: Canonical Correlations, Eigenvalues, and Likelihood Tests

Figure 20.2: Multivariate Statistics and Approximate F Tests

Figure 20.3: Standardized Canonical Coefficients from the CANCORR Procedure

Figure 20.4: Canonical Structure Correlations from the CANCORR Procedure

Chapter 21: The CANDISC Procedure

Figure 21.1: Summary Information

Figure 21.2: MANOVA and Multivariate Tests

Figure 21.3: Canonical Correlations

Figure 21.4: Likelihood Ratio Test

Figure 21.5: Raw Canonical Coefficients

Figure 21.6: Class Means for Canonical Variables

Figure 21.7: Plot of First Two Canonical Variables