Why doesn't Stata allow predicting stdf after using reg y x , vce(robust)?

[Giuseppe Cascarino]

I'm trying to run a Stata code like the following:

reg y x, vce(robust)
predict stfe, stdf

in order to get the standard error of the forecast. However, I get the error

option stdf not allowed after robust estimation

I think that there is nothing preventing from doing this analytically, especially since it is possible to calculate the standard error of the prediction with predict varname, stdp.
See for example Baum (2006) An introduction to modern econometrics using Stata, p.104: once the robust vce matrix has been calculated in order to get the standard error of the prediction, calculating the standard error of the forecast seems a trivial additional step.
Am I missing some reason for which the calculation is not possible, or is it just not implemented in Stata?

[Giuseppe Cascarino]

I guess I found why...
http://www.statalist.org/forums/foru...tandard-errors