Hi! I have a big problem. I am estimating a meta frontier model in which I have 5 regions. All good when I have the results of the region's stochastic frontier, but the problem comes when I am estimating the meta frontier, where the results of the MTE are higher than the technical efficiency. I am using frontier comand, but I do not know what to do. This situation happened with 4 of the 5 regions. Some of my STATA code is this:
frontier y x1 x2 x11 if region==1, distribution(hnormal) uhet(z1 i.z2 i.z3 i.z4 i.z5 i.z6 i.z7)
predict EF8_1 if region==1, te
predict xb8_1 if region==1, xb
and for the meta frontier a code:
frontier pred2 x1 x1 x2 x11, distribution(hnormal) uhet(distance height temp_1 precipitation_1)
predict meta, te
frontier y x1 x2 x11 if region==1, distribution(hnormal) uhet(z1 i.z2 i.z3 i.z4 i.z5 i.z6 i.z7)
predict EF8_1 if region==1, te
predict xb8_1 if region==1, xb
and for the meta frontier a code:
frontier pred2 x1 x1 x2 x11, distribution(hnormal) uhet(distance height temp_1 precipitation_1)
predict meta, te
