STOCHASTIC FRONTIER ANALYSIS OF PRODUCTION FUNCTION AND COST FUNCTION ESTIMATION METHODS. STUDY OF EFFICIENCY AT INDUSTRY LEVEL

This paper analyses the efficiency or productivity at the level of a production unit, but also at industry level, resorting for this purpose to both parametric and non-parametric techniques. Cost function model specifications are described herein, considering that the technical inefficiency effects determine the companies to operate below the production stochastic frontier.<br> Also, the estimations of the production frontier are rendered by means of Cobb-Douglas, C.E.S. and translog functions, evidencing that the latter is the most flexible of them. On the other hand, in many respects, C.E.S. production function is more appropriate to reality as compared to Cobb-Douglas function. At the same time, the estimation of C.E.S. function parameters is more difficult, Cobb-Douglas function being, in this respect, preferred.<br> The last part of the paper is consecrated to the study of efficiency at industry level and to conclusions. The Input variables used within the analysis are: fixed assets, inventories, number of employees, while the Output ones are: operating revenues and net profit.<Br> As for the conclusions, the results reveal that the efficiency level for several companies is quite low. Therefore, a deeper interest should be manifested in order to increase the efficiency level in the construction industry.