基于2004-2012年间全国31个省份的面板数据,同时使用随机前沿分析法(SFA)及数据包络分析法(DEA),实证分析中国各区域研发投入产出的有效性。结果表明:(1)使用SFA法测度,中国省际研发效率随时间呈稳定上升趋势,研发活动存在效率损失,劳动者素质的提高对于推动研发效率更为重要;而使用DEA法测度,中国省际研发效率呈波动上升趋势。(2)两种方法测度出的研发效率值总体上具有方向一致性,DEA法比SFA法测度出的效率值略大,并且两种方法所得效率值之间存在着显著正相关关系。(3)中国研发效率呈现区域不平衡,总体呈现出东部〉西部〉中部的格局。
Based on the panel data of 31 provinces in China during 2004 to 2012,an empirical analysis is made on efficiency of input and output of regional RD based on SFA and DEA methods. The results show that:( 1) By using SFA method,the provincial RD efficiency of China shows a steady upward trend as the time goes; efficiency loss can be found in RD activities,and improving the quality of workers is more important for the promotion of RD efficiency; however,by using DEA method,the provincial RD efficiency shows an upward trend with fluctuations.( 2) the RD efficiency values measured by the two methods have the same trend in general,but the value measured by DEA is slightly larger than the ones estimated by using SFA. Meanwhile,there is a positive correlation between the efficiency values.( 3) China's RD efficiency shows regional imbalances,with the efficiency of eastern region is higher than western region,and the efficiency of western region is higher than central region.