The Multivariate analysis module is useful for exploratory analysis of multivariate quantitative data. Further, it allows you to perform the principal components analysis. Multivariate data (formally a random sample with vector-valued observations) arise as a result of simultaneous measurement of several (m) variables on the same unit. For instance, several physical and/or chemical properties of one sample can be measured, several linear measurements can be taken on the same piece of product, or there might be several characteristics for any employee available.

Multivariate Analysis - Pdf manual

Text outputs: Summary statistics Mean vector Correlation matrix Covariance matrix Explained variability Principal component Eigenvectors Loadings Robust M-estimates Mahalanobis distance, MD Transformed data

Plots: Screeplot Loadings plot Principal component plot Biplot Multivariate normality plot Symmetry plot Andrews curves Robust Mahalanobis distance

Biplot - Powerful classification and visualization tool

Andrews curves help to identify different data

Robust Mahalanobis distance reveals multivariate outliers