摘要：This paper uses two different techniques to measure efficiency in a panel of two hundred multi-product Kansas farms from the period 1984 to 2004. The non-parametric linear programming technique, Data Envelopment Analysis (DEA); and a parametric stochastic frontier approach, a translog production function, are used to calculate and then compare efficiencies measures including pure efficiency, technical, scale and allocative. Production, financial, and demographic variables are used to identify and quantify the causes of inefficiencies. We expect to find variables such as hours of family labor, owned versus rented farm, intensity of production, farmers´ risk attitudes and farmers´ age, significant in explaining farm inefficiency levels in Kansas farms.
