| 摘要: |
| 为了解我国牧区牧业生产增长潜力，利用标准化的220个主要牧区牧场数据构建DEA模型和Tobit、CLAD模型，对牧业生产效率及政府培训、合作社、土地流转、气候带、经营方式等影响因素进行了研究。结果表明：1）我国牧区牧业生产技术效率和规模效率总体偏低，提升空间大，在不同气候带和经营模式下分布差异显著；2）经条件矩检验和LM统计量检验发现，传统Tobit模型扰动项的正态性和同方差性假设在一般情况下难以满足，选用CLAD模型分析影响因素更稳健；3）参加政府培训与合作社，优化从业人员年龄结构，提升从业人员学历、专业化程度、远行距离，有助于生产效率的提高。 |
| 关键词: 畜牧业 技术效率 归并最小绝对离差法(CLAD) 数据包络分析(DEA) |
| DOI：10.11841/j.issn.1007-4333.2017.04.19 |
| 投稿时间：2016-07-08 |
| 基金项目:国家社会科学基金重大项目（11&ZD009） |
|
| Study of the efficiency of livestock production and its influence factors in Chinese main pastoral area: Based on the two-phase modeling of DEA-CLAD |
|
ZHANG Xiaomin1, ZHANG Bingyun2, CHEN Xiaoyu3, ZHANG Zhenghe1
|
| (1.College of Economics and Management, China Agricultural University, Beijing 100083, China;2.College of Economics, Inner Mongolia University of Finance and Economics, Hohhot 010070, China;3.The Communist Party and Administration Office, China Agricultural University, Beijing 100083, China) |
| Abstract: |
| In order to explore the growth potential of livestock productivity in Chinese pastoral area,The survey data of 220 ranch from the main pastoral area in China were standardized and the livestock production technical efficiency was measured and the effects of government training,cooperatives,land transfer,climatic zone distribution were analyzed and DEA,Tobit and CLAD models were constructed,respectively.The results showed:1) The technical efficiency and scale efficiency as a whole was relatively low and has a large room for improvement.Climatic zone and breeding mode showed a significant different on productivity levels;2) CLAD model was better than Tobit model by the results of Conditional Moments Test and Lagrange Multiplier (LM) Test;3) Participating the government training and cooperatives,optimizing the age structure,improving education level and degree of specialization and travel could improve the efficiency of livestock production. |
| Key words: livestock production technical efficiency censored least absolute deviations(CLAD) data envelopment analysis(DEA) |
