Creating Output Data Sets

You can specify whether to create an observationwise statistics data set. This data set contains the sum of squares and cross-products.

You can also choose to include these statistics in the output data set:

predicted values
press statistic, which is the ith residual divided by , where h is the leverage, and where the model has been refit without the ith observation
residual
studentized residuals, which are the residuals divided by their standard errors
studentized residual with current observation removed
Cook’s D influence
the standard influence of observation on covariance of betas
the standard influence of an observation on predicted value (called DFFITS)
leverage