Editors' introduction

In this paper we consider a problem of estimating K common coefficients in two regression models. Assuming that the disturbances of the two equations have different variances, it is shown that, under some general conditions, it is possible to combine information on both the equations to obtain estimators which are more efficient than the one based on just one of the two equations.     

Recovering the production technology from the cost function

An alternative technique for recovering the technology from the cost function is investigated. The recovery method uses an optimization problem with a finite number of constraints that, under appropriate conditions, has, the equations of Shephard's lemma among the first order conditions. Hence, a justification is provided for the technique of recovering the production function from the cost function by solving the system of equations generated by Shephard's lemma. The alternative technique also recovers the output distance function from the cost function in the multioutput case.The editor for this paper was R. Robert Russell.