Empirical investigation of the real input-output relation in agricultural production
New Medit, vol 11, n.4, (December 2012), pp. 16-22
Jel classification: C14, Q12, C67
This study describes the real nature of the input-output relation by combining parametric and non-parametric methods. Data Envelopment Analysis is applied as complement to regression analysis in order to overcome the drawback of the parametric approach, which assumes that all units operate efficiently. By using data from a sample of 365 farms, technical efficiency is estimated; farms are then divided into production function models, which correspond to different efficiency levels. The results show that the parameters of the whole data set and those of the efficient group differ substantially, indicating that the existence of inefficiency leads to inconsistent estimations.
Data Envelopment Analysis, regression analysis, input-output relation, production function