Estimating Production Function under Endogeneity: A Model Based on Data Envelopment Analysis

Endogeneity and its impact on estimating economic models can be seen in many economic studies. Data envelopment analysis is one of the most common non-parametric methods in which different axioms are used to estimate the production function (efficient frontier). However, the issue of endogeneity and its impact on estimating the efficient frontier is less considered. Cordero et al (2016) indicated that standard models of data envelopment analysis do not perform well in the presence of positive and high endogeneity. In this article, a model based on relaxing convexity axiom is presented in which the Cobb-Douglas function is considered as a real production function. Then, the efficiency of the proposed model is compared with the standard models of the data envelopment analysis and Cobb-Douglas function under positive and high endogeneity. The results show that the proposed model outperforms the counterparts. In addition, by comparing the models in different modes of return to scale, it is observed that the type of return to scale is also effective in determining the efficiency and efficiency of the proposed model compared to its economic counterpart.