%0 Journal Article %T 基于时滞效应的青岛市两阶段科技投入与产出互动关系<br>Interactive relationship between two stages of technology input and output in Qingdao city based on time delay effect %A 董兴林 %A 齐欣< %A br> %A DONG Xing-lin %A QI Xin %J 山东大学学报（理学版） %D 2018 %R 10.6040/j.issn.1671-9352.0.2017.573 %X 摘要： 科技投入是科技产出的重要源泉。近年来,我国科技发达的地区无一不保持着较高的科技投入强度,但科技投入对科技产出的推动作用不仅取决于科技投入的多寡,还取决于科技投入的时间与方向。为了更有效地分析青岛市科技投入对于科技产出的作用,有必要将科技投入对科技产出的影响进行动态定量分析。考虑科技投入到产出的时滞性,应用VAR(vector autoregression)模型,科学选取指标,采用平稳性检验、Granger因果检验、协整检验、脉冲响应函数和方差分解对青岛市2004—2015年间科技投入与产出的数据进行动态关系的实证研究。结果表明:在知识转化阶段,科技投入到科技产出的时滞期为2年;在成果转化阶段时滞期为0.5年;科技投入与科技产出相互影响以及R&D全时人员对科技产出的正向影响显著。<br>Abstract: Investment in science and technology is an important source of output of science and technology. In recent years, all areas with developed science and technology in our country have consistently maintained a high intensity of investment in science and technology. However, the role of science and technology in promoting scientific and technological output depends not only on the amount of investment in science and technology, but also on the timing and direction of investment in science and technology. In order to more effectively analyze the effect of science and technology investment on science and technology output in Qingdao, it is necessary to dynamically and quantitatively analyze the impact of science and technology input on science and technology output. Considering the lag effect of input and output, the VAR model is used to select the scientific indicators, and the results are analyzed by stationary test, Granger causality test, Johansen test impulse response function and variance decomposition using science and technology data in Qingdao from 2004 to 2015. The results show that in the stage of knowledge transformation, the time lag of science and technology input to output is two years, and the time lag is 0.5 years in the transformation stage of transformation. Science and technology investment and technology output influence each other. R&D full time staff impacts science and technology output significantly %K 科技产出 %K 两阶段 %K 时滞效应 %K 向量自回归模型 %K 科技投入 %K < %K br> %K time lag effect %K science and technology output %K science and technology input %K two stages %K vector autoregression(VAR)model %U http://lxbwk.njournal.sdu.edu.cn/CN/10.6040/j.issn.1671-9352.0.2017.573