Using Stepwise Discriminant Analysis as a Post-Hoc Procedure to a Significant Hotelling's T2 Control Chart

• Activity Number: 539
• Type: Contributed
• Date/Time: Wednesday, August 4, 2010 : 10:30 AM to 12:20 PM
• Sponsor: Section on Quality and Productivity
• Author(s): Emily Pollock*+ and Jay R. Schaffer
• Companies: University of Northern Colorado and University of Northern Colorado
• Keywords: Multivariate ; Control chart ; Hotelling's T2 ; discriminant analysis

Abstract

Finding post-hoc procedures that effectively identify the cause of out-of-control signals for multivariate control charts has proven difficult for researchers. Due to its multivariate nature, it has been suggested that stepwise discriminant analysis may accurately identify which variables are the most important in explaining shifts in a process. Methods and results are discussed for using stepwise discriminant analysis as a post-hoc procedure to a significant Hotelling's T2 control chart for a variety of situations. Situations include: sample size, number of dependent variables, the correlations among dependent variables, effect size, and alpha level.