Ridge regression is a form of regularized regression that allows for numerous, potentially correlated, predictors and shrinks them using a common variance component model. The process computes Best Linear Unbiased Predictions (BLUPs) of the responses based on this mixed model. Computations are performed using SAS/STAT PROC MIXED.
Refer to the SAS PROC MIXED documentation for more information.
As always, it is not easy to tell beforehand which predictive model best fits your data. You should, therefore, plan to run your data through several, if not all, of the predictive models to find out which model works best. The Cross Validation Model Comparison process is especially useful for this task. See Cross Validation Model Comparison for more details.
One wide Input Data Set is needed to run this process. This data set contains all of the numeric and other data to be analyzed. Data must be in the wide format. Genetic marker data is likely in this form already, but any data that are in tall form must be converted to the wide format. The Transpose Rectangular process can be used to convert the tall data set and its accompanying Experimental Design Data Set (EDDS) to wide form.
The adsl_dii.sas7bdat data set, used in the following example, consists of 906 rows of individuals with 382 columns corresponding to data on these individuals. It was generated from the original nicardipine ADSL data set described in Nicardipine and is included with JMP Clinical. This data set is partially shown below.
For detailed information about the files and data sets used or created by JMP Life Sciences software, see Files and Data Sets.
The output generated by this process is summarized in a Tabbed report. Refer to the Ridge Regression output documentation for detailed descriptions and guides to interpreting your results.