I can see why someone might prefer it that way. I think it is in fact one of the options when installing SAS. I personally don't prefer it. I like being able to double-click and open into the SAS environment. One advantage is that that sets the folder in which the .sas file is stored as the work directory. ... View more

Yes, you are right, I was sloppy in creating the datasets Have and Want by hand. I like your solution a lot and will be keeping it. So the thread contains three excellent and very different solutions. ... View more

Hi (and particularly @StatDave 😞   Would it be possible to generate a bootstrap estimate of the residual variance?  Assume the range of X values is uniform. Assume the parameter estimates are normally distributed, with mean = the reported estimate and standard deviation = the reported standard error. Draw at random with replacement 100 values of X, and 1000 3-ples of the parameters (this is where a big assumption comes in - that the parameter estimates are uncorrelated). From these, calculate 100 x 1000 Y values. Last step, refit the equation at a range of (X,Y) values covering the interval, making sure to include X values of interest with Y set to missing. That should give predicted Y values with standard errors and confidence intervals, and I am sure quantiles could be calculated given the residual error.   Thoughts?   SteveDenham ... View more

Combining PGStats and StatsMan answers gives a full answer to my immediate question, but Stats Dave provided a great answer to my general underlying need. I did not realize that GLIMMIX and GEE had the link function capability Stats Dave pointed out, and I also appreciate the suggestion that maybe a repeated measure approach would be appropriate. Since only one response gets to be marked as the solution, I will go with StatsDave, but readers who want to know why PROC NLMIXED generated an estimate of u, and why I shouldn't even have included u in the parms statement will want to look at PGStats and StatsMan answers. ... View more