STAT 520 Multivariate Analysis for Data Mining

Concept-based introduction to multivariate analysis, useful for data mining and predictive modeling, with emphasis given to interpreting output and checking model assumptions using one of the standard statistical package. Topics may include: multivariate normal distribution, simultaneous inferences, one- and two-way MANOVA, multivariate multiple regression and ANACOVA, correlation, principle component and facor analysis, discriminant analysis, cluster analysis and multidimensional scaling, path analysis, structural equation modeling, and longitudinal data analysis.

Credits

4

Prerequisite

Two semesters of applied statistics (such as STAT 104/STAT 453, STAT 200/STAT 201, or STAT 215/STAT 216), or two semesters of statistics approved by advisor, or permission of department chair.

General Education

Offered

  • Fall