参见help frontier_postestimation，用predict [type] newvar [if] [in] [, statistic]

Syntax for predict

        predict [type] newvar [if] [in] [, statistic]

        predict [type] {stub*|newvar_xb newvar_v newvar_u} [if] [in] , scores

    statistic    description
    --------------------------------------------------------------------------------------------
    Main
      xb         linear prediction; the default
      stdp       standard error of the prediction
      u          estimates of minus the natural log of the technical efficiency via E(u|e)
      m          estimates of minus the natural log of the technical efficiency via M(u|e)
      te         estimates of the technical efficiency via E{exp(-su)|e}
                   s =  1, for production functions
                   s = -1, for cost functions
    --------------------------------------------------------------------------------------------
    These statistics are available both in and out of sample; type predict...if e(sample)... if
      wanted only for the estimation sample.

Options for predict

        +------+
    ----+ Main +--------------------------------------------------------------------------------

    xb, the default, calculates the linear prediction.

    stdp calculates the standard error of the linear prediction.

    u produces estimates of minus the natural log of the technical efficiency via E(u|e).

    m produces estimates of minus the natural log of the technical efficiency via M(u|e).

    te produces estimates of the technical efficiency via E{exp(-su)|e}.

    scores calculates equation-level score variables.

        The first new variable will contain the derivative of the log likelihood with respect to
        the regression equation.

        The second new variable will contain the derivative of the log likelihood with respect
        to the second equation (lnsig2v).

        The third new variable will contain the derivative of the log likelihood with respect to
        the third equation (lnsig2u).