1. Home
  2. Multivariate and cluster analysis

Multivariate and cluster analysis

Several standard multivariate methods are provided by Genstat directives. These include methods that analyse data in the form of units-by-variates, and methods that use a similarity or distance matrix.

The following directives carry out standard multivariate analyses:

    CVA canonical variates analysis
    FCA factor analysis
MDS non-metric multidimensional scaling
    PCP principal components analysis
    PCO principal coordinates analysis
    ROTATE Procrustes rotation

Other directives and procedures are available to process results from multivariate analyses:

ADDPOINTS    adds points for new objects to a PCO
CVAPLOT plots the mean and unit scores from a canonical variates analysis
CVASCORES calculates scores for individual units in canonical variates analysis
DBIPLOT plots a biplot from an analysis by PCP, CVA or PCO
DMST gives a high resolution plot of an ordination with minimum spanning tree
FACROTATE rotates factor loadings from a PCP, CVA or FCA     ADDPOINTS  
LRVSCREE prints a scree diagram and/or a difference table of latent roots
 PCORELATE relates principal coordinates to original data variables

The following directives are used for hierarchical or non-hierarchical cluster analysis:

CLUSTER non-hierarchical clustering from a data matrix
    FSIMILARITY forms a similarity matrix or a between-group similarity matrix from a units-by-variates data matrix
    HREDUCE forms a reduced similarity matrix (by groups)
    HCLUSTER hierarchical cluster analysis from a similarity matrix

Other directives and procedures that process the results from hierarchical cluster analyses are:

DDENDROGRAM draws dendrograms with control over structure and style
DCLUSTERLABELS labels clusters in a single-page dendrogram plotted by DDENDROGRAM
HBOOTSTRAP performs bootstrap analyses to assess the reliability of clusters from hierarchical cluster analysis
HCOMPAREGROUPINGS compares groupings generated, for example, from cluster analyses
    HDISPLAY displays results associated with hierarchical clustering
HFAMALGAMATIONS forms an amalgamations
HFCLUSTERS forms a set of clusters from an amalgamations matrix
    HLIST lists a data matrix in abbreviated form
HPCLUSTERS prints a set of clusters
    HSUMMARIZE summarizes data variates by clusters

Other multivariate techniques are provided by procedures in the Library:

    AMMI allows exploratory analysis of genotype × environment interactions
    BCLASSIFICATION constructs a classification tree
    BCDISPLAY displays a classification tree
    BCIDENTIFY identifies specimens using a classification tree
    BCKEEP saves information from a classification tree
    BCVALUES forms values for nodes of a classification tree
    BCFOREST constructs a random classification forest
    BCFDISPLAY displays information about a random classification forest
    BCFIDENTIFY identifies specimens using a random classification forest
    BIPLOT produces a biplot from a set of variates
    BKEY constructs an identification key
    BKDISPLAY displays an identification key
    BKIDENTIFY identifies specimens using a key
    BKKEEP saves information from an identification key
    CANCORRELATION does canonical correlation analysis
    CCA performs canonical correspondence analysis
    CRBIPLOT plots correlation or distance biplots after CCA or RDA
    CRTRIPLOT plots ordination biplots or triplots after CCA or RDA
    CINTERACTION clusters rows and columns of a two-way interaction table
    CLASSIFY obtains a starting classification for non-hierarchical clustering
    CONVEXHULL finds the points of a single or a full peel of convex-hulls
    CORANALYSIS does correspondence analysis, or reciprocal averaging
    MCORANALYSIS does multiple correspondence analysis
    CABIPLOT plots results from correspondence analysis or multiple correspondence analysis
    DISCRIMINATE performs discriminant analysis
    SDISCRIMINATE selects the best set of variates to discriminate between groups
    QDISCRIMINATE performs quadratic discrimination between groups i.e. allowing for different variance-covariance matrices
    DPARALLEL displays multivariate data using parallel coordinates
    GESTABILITY calculates stability coefficients for genotype-by-environment data
    GGEBIPLOT plots displays to assess genotype + genotype-by-environment variation
    GENPROCRUSTES performs a generalized Procrustes analysis
    IDENTIFY identifies an unknown specimen from a defined set of objects
KNEARESTNEIGHBOURS classifies items or predicts their responses by examining their k nearest neighbours
    MANOVA performs multivariate analysis of variance and covariance
    MANTEL assesses the association between similarity matrices
    MULTMISSING estimates missing values for units in a multivariate data set
    MVAOD does an analysis of distance of multivariate data
    NORMTEST performs tests of univariate and/or multivariate normality
    OPLS performs orthogonal partial least squares regression
    PCOPROCRUSTES performs a multiple Procrustes analysis
    PLS fits a partial least squares regression model
    RDA performs redundancy analysis
    RIDGE produces ridge regression and principal component regression analyses
    LRIDGE does logistic ridge regression
    RLFUNCTIONAL fits a linear functional relationship model
    RMULTIVARIATE performs multivariate linear regression with accumulated testing of terms
    ROBSSPM forms robust estimates of sum-of-squares-and-products matrices
    SAGRAPES produces statistics and graphs for checking sensory panel performance
    SKEWSYMMETRY provides an analysis of skew-symmetry for an asymmetric matrix
Updated on March 11, 2022

Was this article helpful?