基于GM-DEA-Tobit模型的农业新质生产力评估与优化研究——以湖南省14地市州为例

Study on the Evaluation and Optimization of Agricultural New-Quality Productive Forces Based on the GM-DEA-Tobit Model—A Case Study of 14 Prefecture-Level Cities and Autonomous Prefectures in Hunan Province

摘要: 本文基于DEA-Malmquist、Tobit模型及灰色预测(GM)，旨在评估与优化湖南省农业的新质生产力。研究2017~2023年湖南省14个地市州面板数据，构建“效率测度–影响因素–趋势预判”三阶段框架。运用DEA-Malmquist模型测度农业全要素生产率(TFP)及分解项(技术效率、技术进步)时空演变；通过Tobit模型识别影响TFP关键因素，考察农业科技研发投入等变量作用；用灰色预测(GM)模型预测2024~2029年TFP及分解项趋势。发现研究期内湖南农业TFP年均降0.9%、2021年后回升，区域差异显著；农业研发投入、机械化配置正向影响TFP，能源利用效率负向影响；预计2024~2029年TFP年均增2.65%，技术进步驱动但技术效率存隐忧，提出强化创新等建议。

Abstract: This paper is based on the DEA-Malmquist, Tobit model, and grey prediction (GM), aiming to evaluate and optimize the new-quality agricultural productivity in Hunan Province. It studies the panel data of 14 prefecture-level cities in Hunan Province from 2017 to 2023 and constructs a three-stage framework of “efficiency measurement-influencing factor-trend prediction”. The DEA-Malmquist model is used to measure the spatial-temporal evolution of agricultural total factor productivity (TFP) and its decomposition items (technical efficiency, technological progress); the Tobit model is adopted to identify the key factors influencing TFP and examine the roles of variables such as agricultural science and technology R & D investment; and the grey prediction (GM) model is utilized to predict the trends of TFP and its decomposition items from 2024 to 2029. It is found that during the research period, the average annual TFP of agriculture in Hunan decreased by 0.9% and rebounded after 2021, with significant regional differences; agricultural R & D investment and mechanization configuration have a positive impact on TFP, while energy utilization efficiency has a negative impact; it is predicted that the TFP will increase at an average annual rate of 2.65% from 2024 to 2029, driven by technological progress but with concerns about technical efficiency, and suggestions such as strengthening innovation are put forward.