Description

The main objective of this course is on the application of multivariate statistical methodologies to research data. The topics include multivariate normal distribution; its properties and inference; multivariate analysis of variance (MANOVA), analysis of covariance (MANCOVA) and multivariate regression; multivariate models for repeated measures analysis; principal component analysis, factor analysis, discriminant analysis, cluster analysis and structural equation modeling.

This is an advanced course in multivariate statistical analysis methodology with applications. The choice of statistical software is Statistical Analysis Software (SAS). There will be an extensive use of SAS’s Interactive Matrix Language (IML) procedure as well as procedures such as PRINCOMP, FACTOR, CALIS, DISCRIM in this course.

Prerequisite

• Completion of Advanced Course in Biostatistics.
• Knowledge of SAS to create and manage data sets as well as experience using Generalized Linear Model procedures in SAS.

Learning Outcomes

• Have a solid understanding of the statistical methodologies that are appropriate to analyze data that are multivariate in nature
• Use appropriate procedures in the SAS system to conduct methodologies outlined above and interpret the analytical results.

Topics of Study

The course focuses on the following statistical methodologies with an introduction to matrix algebra:

• Introduction to Matrices and Linear Algebra
• Introduction to Multivariate Distribution with Examples
• Multivariate Analysis of Variance (MANOVA)
• Multivariate Linear Models
• Principal Component Analysis
• Factor Analysis
• Discriminant Analysis
• Cluster Analysis
• Structural Equation Modeling