OVERALS

OVERALS is available in Sampling and Testing.

OVERALS performs nonlinear canonical correlation analysis on two or more sets of variables. Variables can have different optimal scaling levels, and no assumptions are made about the distribution of the variables or the linearity of the relationships.

OVERALS VARIABLES=varlist (max)

/ANALYSIS=varlist[({ORDI**})] 
                   {SNOM  }
                   {MNOM  }
                   {NUME  }

/SETS= n (# of vars in set 1, ..., # of vars in set n)

[/NOBSERVATIONS=value]

[/DIMENSION={2**  }] 
            {value}

[/INITIAL={NUMERICAL**}] 
          {RANDOM     }

[/MAXITER={100**}] 
          {value}

[/CONVERGENCE={.00001**}] 
              {value   }

[/PRINT=[DEFAULT] [FREQ**] [QUANT] [CENTROID**]
        [HISTORY] [WEIGHTS**]
        [OBJECT] [FIT] [NONE]]

[/PLOT=[NDIM=({1    ,2    }**)] 
              {value,value}
              {ALL  ,MAX  }
       [DEFAULT[(n)]] [OBJECT**[(varlist)][(n)]] 
       [QUANT[(varlist)][(n)]] [LOADINGS**[(n)]] 
       [TRANS[(varlist)]] [CENTROID[(varlist)][(n)]]
       [NONE]]

[/SAVE=[rootname][(value)]]

[/MATRIX=OUT({*                  })]
             {'savfile'|'dataset'}

**Default if the subcommand or keyword is omitted.

This command reads the active dataset and causes execution of any pending commands. See the topic Command Order for more information.

Syntax for the OVERALS command can be generated from the Nonlinear Canonical Correlation Analysis (OVERALS) dialog.

Example

OVERALS VARIABLES=PRETEST1 PRETEST2 POSTEST1 POSTEST2(20)
SES(5) SCHOOL(3)
/ANALYSIS=PRETEST1 TO POSTEST2 (NUME) SES (ORDI) SCHOOL (SNOM)
/SETS=3(2,2,2).