Principal Components and Classification Analysis Button

Click the button to display the Advanced Principal Components and Classification Analysis (Startup Panel). The PCCA module is useful for mapping variables and cases (observations) into the factor space (dimensions) computed from a set of analysis variables and cases; for example, you could compute a factor space (dimensions) from a set of key variables and cases that represent well the area or phenomenon of interest, and then apply the dimensions thus derived to a new set of cases and variables.

The PCCA module computes principal components for large numbers of variables, and a wide range of associated statistics. You can specify supplementary variables and supplementary cases (observations), which will not be used for the extraction of principal components, but can be mapped into the coordinate system (factor structure) determined from the variables and cases selected for the analysis. Supplementary variables and cases will be included in all results tables and graphs, for example in the scatterplot of factor loadings, where supplementary variables and cases are labeled and identified by different point markers. The PCCA module computes all standard results statistics, including factor coordinates of variables and cases (and supplementary variables and cases), contributions of variables and cases to the variance, eigenvalues and eigenvectors, factor scores, factor score coefficients, cosine-squares, etc. A large number of 2D and 3D plots are available including plots of eigenvalues, simple scatterplots, and, of course, plots of the factor coordinates for variables and cases, including supplementary variables and cases.

Factor analysis and similar analyses can also be performed with various other STATISTICA modules, such as Factor Analysis, Structural Equation Modeling and Path Analysis (SEPATH), Correspondence Analysis, and Multidimensional Scaling.