I would like to use the now out-of-date Stata program "regpred2".... it
still uses the old graphics language. It still works, but the graphics are
not as nice as under version 8.
For those unfamiliar with it, regpred2 displays and plots the predicted
values and their confidence intervals from a linear regression for
selected values of a continuous xvar. It can be thought of as a more
powerful version of "twoway lfitci". With regpred2 you can control for
more than one independent variable.
