MATH-652 Applied Multivariate Analysis
Offered academic year 2014-2015
Multivariate data analysis refers to the analysis of several response variables simultaneously. The overarching goal of this course, therefore, is to provide students with the knowledge necessary to make proper interpretations and select appropriate techniques for analyzing several correlated outcome variables concurrently. Course topics include review of matrix algebra and random vectors, sample geometry and random sampling, the multivariate normal distribution, inferences about a mean vector, comparisons of several multivariate means via multivariate analysis of variance (MANOVA), and multivariate linear and multiple regression models. The course also covers principal component analysis, factor analysis and inference for structured covariance matrices; canonical correlation analysis; and discrimination and classification. Analyzing all the outcomes simultaneously is more efficient and can lead to more power compared to evaluating each outcome separately. These tools are!
used in a broad array of applications including biostatistics, psychometrics, behavioral sciences, business and finance, social sciences, quality control and process control, etc.
Students gain theoretical comprehension of the above material, and complement this understanding with applied computational work using SAS and/or R, including a data analysis project and class presentation. This is an advanced course for graduate students in statistics, and graduate students from applied fields with sufficient background in statistics and linear algebra.
Required Text: R.A. Johnson, and D.W. Wichern (2007) Applied Multivariate Statistical Analysis, sixth edition.
Must be enrolled in one of the following Levels:
MN or MC Graduate
Must be enrolled in one of the following Majors:
Mathematics and Statistics
Prerequisites: Math-503 or equivalent
Other academic years
There is information about this course number in other academic years: