Principal Components and Classification Analysis

Complete implementation of Principal Components Analysis for classification of variables and cases (observations); Statistica computes Principal Components for selected variables and observations, and applies the resultant factor structure to map supplementary variables and observations into the same space (see also Jambu, 1991 for additional details). These computations can be based on the covariance matrix or correlation matrix. Various results spreadsheets and graphs are available to review the mapping of the variables and observations.

General

Element Name Description
Detail of computed results reported Specifies the level of computed results reported. At the Minimal level of detail, Statistica will report the Eigenvalues and factor structure for the variables and supplementary variables; at the Comprehensive level of detail, Statistica will also create plots of the factor coordinates, and report the factor score coefficients, communalities, and other results that depend on the specific analysis settings; if All results is requested, various descriptive statistics are also reported. Case statistics (for observations) can be computed as an option.
MD Deletion Missing data can be deleted Casewise, or missing data can be substituted by the means for the respective variables.
Analyzes matrix Specifies whether the analyses are to be computed based on the correlation matrix (i.e., the data will be centered and scaled), or the covariance matrix (data will be centered, but not scaled).
Creates variances as Select how to estimate sigma from the Sums of Squares (SS); there are two ways in which the variances can be computed: as the sums of squares of deviations divided by N (the valid number of cases) or N-1.
Creates case statistics Creates statistics (contributions, cosine-squares, etc.) and plots for cases (observations).
Generates data source, if N for input less than Generates a data source for further analyses with other Data Miner nodes if the input data source has fewer than k observations, as specified in this edit field; note that parameter k (number of observations) will be evaluated against the number of observations in the input data source, not the number of valid or selected observations.

Plots of Coordinates

Element Name Description
Max. n of factor in plots Specifies the maximum number of factors to depict in plots of factor coordinates.
Vectors to origin Plots vectors from the origin to the factor coordinates in plots of factor coordinates.
Draw Unit Circle Plots the unit circle, encircling the factor coordinates, in plots of factor coordinates.
Name variables in plots Specifies how to label the variables in 2D plots of factor coordinates for variables.
Name cases in plots Specifies how to label the cases in 2D plots of factor coordinates for cases; case names are only available if they exist in the input data file, and Grouping labels are only available if a (optional) Grouping variable (Learning variable) is specified.
Plots cases with cos^2 >= Specifies a cutoff value for cases to be shown in plots of cases; only cases with contributions (cosine-squared) greater than this value will be shown.