Dear Statalist:

In the past, when I used -lpoly- to estimate smoothed values over groups, I just created a separate data file for the subgroups and ran it separately. This works fine if you're estimating it by gender, and less if you're estimating by, say, USA states (or states X gender).

As far as I can tell, lpoly doesn't work with -by- or its cousins. There are probably reasons for this, but I'm hoping to run lpoly on about 100 different subgroups and creating different data files is a bit of a mess (if you're wondering: Yes, my data set is exceptionally large). Maybe it would work if I created temporary data files and then merged them together rather than permanently throwing 100 data files in a folder somewhere, but I'm hoping that there's a simpler way to approximate this. Does anyone have any experience making this work?

Thanks,
Jonathan