基于超效率DEA-Malmquist指数的人工智能产业创新效率评价

Research on Innovation Efficiency Evaluation of Artificial Intelligence Industry Based on Super-efficiency DEA-Malmquist Index

本文运用超效率DEA从静态层面测算了2010—2020年我国30个省份的人工智能产业创新效率,并采用Malmquist指数对人工智能产业创新效率的变动情况进行动态分析,对人工智能产业全要素生产率进行分解,分析影响创新效率进步关键指标.研究发现:从整体发展水平看,我国30个省份的人工智能产业创新效率总体呈平缓上升趋势;从地区发展水平看,东、中、西部地区的人工智能产业创新效率存在明显的差异性,东部地区呈现缓慢下降的趋势,但仍继续维持领跑态势,中部地区人工智能产业创新效率总体呈缓慢上升趋势,西部地区人工智能产业创新效率呈波动式变化,但变化不大且始终低于平均效率;从全要素生产率来看,我国人工智能产业全要素生产率整体呈波动上升趋势,主要受技术效率和技术进步指数的影响,技术效率的提升得益于规模效率和纯技术效率的增长.

This article used super-efficiency DEA to measure the innovation efficiency of AI industry in 30 provinces of China from 2010 to 2020,and used Malmquist index to analyze the change in innovation efficiency of AI industry dynamically,breaking down the total factor productivity of the AI industry and analyzing key indicators of progress affecting innovation efficiency.It is found that the innovation efficiency of AI industry in 30 provinces of China is on the rise in the whole level of development and that the innovation efficiency of AI industry is on the rise in the whole level of regional development.There are obvious differences in the innovation efficiency of the AI industry in the eastern Midwestern Sectional Figure Skating Championships.The eastern region shows a slow downward trend,but it continues to maintain its leading position.The innovation efficiency of the AI industry in the central region of China is slowly rising,while that in the Western Region is fluctuating,but the change is small and always below average.From the total factor productivity perspective,the overall total factor productivity of the AI industry in China shows a fluctuating upward trend,which is mainly influenced by the technical efficiency and the technical progress index.The improvement in technical efficiency is due to the increase in scale efficiency and pure technical efficiency.