04-16-2014 08:15 AM
I’m trying to transform variable using box-cox transformation for normality.
This variable is combination of negative, positive and missing values.
My SAS code:
proc transreg data=data ss2 details plots=(transformation(dependent) scatter
model BoxCox(change_tc_wk12 / lambda=-2 -1 -0.5 to 0.5 by 0.05 1 2
alpha=0.01) = identity(x);run;
Log error message:
ERROR: 273 invalid values were encountered while attempting to transform variable change_tc_wk12.
WARNING: Ordinary missing values were found or an UNTIE transformation or the UNTIE= option was specified. The utility of the
hypothesis tests are dubious since one parameter must be estimated for each of these values. If you really want to
do this, ensure that no observations are duplicated -- combine duplicate observations and use a FREQ statement. If
you do not, the parameter count may be too large and the tests overly conservative. However, it is best to avoid
this situation altogether.
NOTE: Plot requests DEPENDENTS SCATTER OBSERVEDBYPREDICTED are ignored.
NOTE: The SAS System stopped processing this step because of errors.
While same code works for another variable containing positive values.
Also is there any procedure to identify possible transformation for particular non normal variable similar to ‘ladder’ command in STATA software which provides all possible transformations.