Dear all,

Your assistance will be greatly appreciated!

I am running a stochastic frontier model with two wave panel data. when i put in the command,
sfpanel y x1 x2..xn, model (tfe) dist(tn) emean (z1...zn), if hh==1, it gives me me this, Warning: only units with more than 1 time occasion will be considered'.

The generated estimates will then be based on only the observations of the first wave and not the entire observations of the two panels. what could be the problem? Are there any other specifications i can use to get the entire observation in the estimates?

when i use the cross section command for the inefficiency effect as:
sfpanel y x1 x2..xn, dist(tn) uhet (z1...zn), if hh==1, then it counts both wave 1 and 2 observations but does not generate the inefficiency effects.