Box-Cox Transformation
Ribbon bar. Select the Data tab. In the Transformations group, click Box-Cox to display the Box-Cox Transformation dialog box.
Classic menus. With a data set open, on the Data menu, select Box-Cox Transformation to display the Box-Cox Transformation dialog box.
See also, Box-Cox Transformation Overview and Technical Notes.
| Option | Description |
|---|---|
| Variables | Displays a standard variable selection dialog box. From the first list, select the variables to be transformed using the Box-Cox method. From the second list, select one or more grouping variables. When one or more grouping variables are selected for the Box-Cox analysis/transformation, the transformation is computed with the goal of transforming the within group distributions to normality with homogenous variances; this option is useful when analyses (example. ANOVA) are to be performed with categorical predictors. |
| Box-Cox transformation | |
| Max. iterations. | Specifies the maximum number of iterations that can be used by the algorithm for finding a suitable Box-Cox transformation for each and every selected variable and on a by group basis (provided you have selected at least one by grouping variable). |
| Min. lambda. | Specifies the minimum value of
 to start with.
|
| Max. lambda. | Specifies the maximum value of
to be tried by the search algorithm.
|
| Epsilon (convergence) | The search for optimal
(and, hence, the best Box-Cox transformation) needs a convergence criteria, which is used to decide whether the algorithm has found a solution with reasonable accuracy. If convergence is achieved before the maximum number of iterations is reached, the algorithm stops. This feature is particularly useful when dealing with many variables and large data sets as it can save you time in producing your Box-Cox models.
|
| Shift variable with minimum<=0 to | You can specify the value of the shift parameter
 of the Box-Cox transformation.
|