Options (MULTIPLE CORRESPONDENCE command)
Discretization. You can use the DISCRETIZATION
subcommand to discretize fractional value variables or to recode
categorical variables.
Missing data. You can specify the treatment of missing data per variable with
the MISSING
subcommand.
Supplementary objects and variables. You can specify objects and variables that you want to treat as supplementary.
Read configuration. MULTIPLE CORRESPONDENCE
can
read a configuration from a file through the CONFIGURATION
subcommand. This configuration can be used
as the initial configuration or as a fixed configuration in which
to fit variables.
Number of dimensions. You can specify how many dimensions MULTIPLE
CORRESPONDENCE
should compute.
Normalization. You can specify one of five different options for normalizing the objects and variables.
Tuning the algorithm. You can control the values of algorithm-tuning parameters with the MAXITER
and CRITITER
subcommands.
Optional output. You can request optional output through the PRINT
subcommand.
Optional plots. You can request a plot of object points, transformation plots per variable, plots of category points per variable, or a joint plot of category points for specified variables. Other plot options include residuals plots, a biplot, and a plot of discrimination measures.
Writing discretized data, transformed data, and object scores. You can write the discretized data, the transformed data, and the object scores to outfiles for use in further analyses.
Saving transformed data and object scores. You can save the transformed variables and the object scores in the working data file.