sfcross fitsstochastic production or cost frontier models; the default is a productionfrontier model.  It provides estimatorsfor the parameters of a linear model with a disturbance that is assumed to be amixture of two components, which have a strictly nonnegative and symmetricdistribution, respectively.  sfcross canfit models in which the nonnegative distribution component (a measurement ofinefficiency) is assumed to be from a half-normal, exponential,truncated-normal or gamma distribution. In the latter case, maximization is performed through maximumsimulated likelihood.

Syntax:sfcross depvar[indepvars] [if] [in] [weight] [, options]

frontier fitsstochastic production or cost frontier models; the default is a productionfrontier model.  It provides estimatorsfor the parameters of a linear model with a disturbance that is assumed to be amixture of two components, which have a strictly nonnegative and symmetricdistribution, respectively.  frontier canfit models in which the nonnegative distribution component (a measurement ofinefficiency) is assumed to be from a half-normal, exponential, ortruncated-normal distribution.  SeeKumbhakar and Lovell (2000) for a detailed introduction to frontier analysis.

Syntax: frontier depvar [indepvars] [if] [in][weight] [, options]

从stata的help的description里就只看到红色的区别，用数据分别跑了一下，两个的结果差别还挺大的，这两个命令到底有什么区别呢？