How to get SSE for predictions using SAS?

Question

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;

Answer 1

I would check the options in PROC glm to see how to get that output. The help function in SAS is very good and you can find what you want with keyword searches or by skimming through the SAS/STAT users guide. But it is easy enough to program it yourself. SAS gives you the regression coefficients and the intercept term. Just right down the regression equation with the coefficients plugged in and the values of the covariates at the point(s) where you want to make the prediction and sum the squared differences..