This course is for JMP users who work with data that have many variables. The course demonstrates various ways to examine high-dimensional data in fewer dimensions, as well as patterns that exist in the data. Methods for unsupervised learning are presented in which relationships between the observations, as well as relationships between the variables, are uncovered. The course also demonstrates various ways of performing supervised learning where the relationships among both the output variables and the input variables are considered. In the course, emphasis is on understanding the results of the analysis and presenting conclusions with graphs.


Before attending this course, you should complete the JMP Software: Statistical Decisions Using ANOVA and Regression course.

This course addresses JMP software.

Delegates will learn how to

  • use principal component analysis to reduce the number of data dimensions
  • use loading plots to understand the relationships between variables
  • interpret principal component scores and perform factor analysis
  • build more stable models by removing collinearity with principal component regression (PCR)
  • identify natural groupings in the data via cluster analysis
  • identify clusters of variables
  • classify observations into groups with discriminant analysis
  • fit complex multivariate predictive models with partial least squares (PLS) regression models


Introduction to Multidimensional Data

  • multidimensional analysis
  • review of matrix algebra

Principal Component Analysis

  • interpretation of principal components
  • finding principal components

Principal Component Regression

  • principal component regression

Factor Analysis

  • factor extraction
  • factor rotation

Cluster Analysis

  • introduction to cluster analysis
  • hierarchical clustering
  • K-means clustering
  • variable clustering

Discriminant Analysis

  • discrimination
  • classification

Partial Least Squares Regression

  • PLS algorithms
  • PLS reports