I have a data set with some predictors, a number of training points, and some test points. I've come up with a model that I believe most accurately represents the training data (throwing out x1 and x2), and have come up with some predicted points.
I'm supposed to produce an SSE estimate for the predicted points, but I'm not sure how to programmatically do this in SAS. I was told to look into the stdp regression measure, but I don't know how to use it to get what I need.
EDIT: I'm a total amateur at SAS, so forgive me if my syntax or usage isn't pretty.
data tridata; infile '\data.dat'; input x1 x2 x3 y; run; proc reg data = tridata; model y=x3 noprint; output out = tridata2 r = resid p = pred; run; /* Check predictions */ proc print data=tridata2; var pred; where y eq .; run;