Analysis of Agricultural Production in China and Measurement of Technical Efficiency Using Copula-Based Stochastic Frontier Model

The purpose of this paper is to evaluate the efficiency of agricultural production in China, analyze China's production function, and identify the primary factor affecting efficiency. Methodologically, we employ the stochastic frontier model, accounting for the interrelation between the two-sided error term (U) and the one-sided inefficiency factor (V). To address this complex relationship, we introduce a recent approach known as the Copula-based stochastic frontier model. After obtaining technical efficiency (TE) values, we classify the 31 provinces in China into three distinct groups using the K-means clustering method, each exhibiting unique characteristics. Subsequently, we conduct a LASSO regression analysis to examine the key factors influencing TE values. Our findings reveal that the most significant factor impacting TE values is the average household size, which refer to the average number of laborers per rural household in each province. Furthermore, we observe a negative correlation between the level of education among the labor force and TE values. We analyze that this negative correlation primarily results from the agricultural sector's limited appeal to individuals with higher levels of education.