GEE prediction interval... Help?

#1
Ok, so I'm a bit stuck here... I used a GEE model, with 2 groups (group as factor) within each group subjects walk at two different walking speeds. Since each subject walks at their own speeds, I used speed as a covariate. I get significant effects of Group and Speed, and a significant intercept. So far, I fully understand all my results. What I would like to do now, is plot the regression for one of the groups, together with it's confidence interval, and it's prediction interval. (I want to do this so that later I can plot all data points of the other group, and see how many are actually outside of the prediction interval of the first group_). I'm guessing this should be possible, since I have all the model coefficients and their standard errors.... however, I have NO clue as to what formula to use.... could anyone point me in some direction? Thanks!
Sjoerd