Research On The Time Lags Effects Of Input-output In National Science,Technology And Innovation System

Innovation input is considered as the material basis of innovation,which plays a vital role in enhancing the national innovation capacity.Countries should increase the quantity of innovation input,at the same time,paying more attention to the utilization efficiency of innovation resources,total quantity and quality of innovation output.The transform process from innovation input to innovation output may be delayed because of the interaction and influence of various factors in the national science,technology and innovation system(NSTIS).The innovation input can't generate innovation output and profits immediately,or there are the multiple-period influence of one-time innovation input on output.Therefore,how to accurately identify the long-term and dynamic mechanism between the innovation input and output,and what policy measures can be taken to enhance the efficiency of the innovation input-output,total quantity and quality of innovation output,that are the significant issues faced and needed to solve in the management of national science,technology and innovation.Scientific and reasonable assessment approach for national science,technology and innovation capabilities contributes to clear comprehension of national science,technology and innovation level,and provides important references for formulating the development strategies and promoting the national innovation level.However,the existing assessment methods of national innovation capability don't consider the time lags effects of innovation input-output,and can't reflect the objective law of historical accumulation.Therefore,improving the accuracy of assessment from new perspective and method is becoming a very more important issue.Based on the theory of management,statistics,system science,econometrics,innovation theory,this paper uses the time lags effects of input-output in the NSTIS and its application in the innovation assessment as the study object to study the formation mechanism,influencing factors,and distribution structure of the time lags effects,as well as construct nonlinear assessment methodology considering the time lags effects by means of the system dynamic model,econometric model,and support vector machines.Firstly,constructing the theoretical framework of the time lags effects of input-output in NSTIS.Define the concepts of the time lags effects of input-output.Determine four major entities: governments,enterprises,research and technological organizations and educational institutions.Put forward the structure model containing four subsystems: innovation foundation subsystem,innovation environment subsystem,innovation carrier subsystem,and innovation output subsystem.Propose index system of NSTIS composed of 38 indicators by frequency statistics,membership analysis,correlation analysis and discrimination analysis.The conceptual framework of NSTIS lays the theoretical research foundation for the time lags effects.Secondly,developing a NSTIS model with the use of system dynamic methodological approach to identify and analyze the time lags in the transform process from innovation input to innovation output.Introduce the causal loop diagram for the model based on conceptual framework and identify the four different feedback loops.Establish the flow diagram of NSTIS of China and simulate the model based on Vensim.Detect the effectiveness of model.By adding pulse function or step function,study the lags variation between the innovation input and output at different levels of pulse or step.Analyze the formation mechanism and influencing factors of time lags in the four feedback loops.Thirdly,establishing a finite distributed lag model to measure the structure and extent of time lags effects of innovation input on output.Test all variables for unit roots by the Augmented Dickey Fuller(ADF)test with the help of EViews.Employ the Almon polynomial method to measure the initial lags length and coefficient.Use the Kalman filter method to adjust the estimated parameters.Fourthly,building the nonlinear assessment model of national innovation capability based on time lags effects and Least Square Support Vector Machine(LS-SVM).In order to verify the accuracy of the model,the model of national innovation capability is verified by G20 member countries as the research object,and the innovation capabilities of G20 member countries are evaluated.Finally,policy implications drawn from empirical results are discussed.We propose some policy recommendations considering the time lags effects of innovation input-output from four aspects: innovation personnel,innovation expenditure,innovation policies and regulations and industry-academia-research collaboration,based on the time lags analysis results and innovation development trajectory of China.Then some policy recommendations are put forward from the four aspects: innovation foundation,innovation environment,innovation carrier and innovation output,based on the national innovation capability assessment results.On the whole,this study contributes to theory-building and policy-making in the following ways.For theory-building,the investigation of the dynamics and time lags properties of the long-run evolution of NSTIS can complement and inform the national innovation system research,and reveal the long-term complex relationships between the innovation input and output to further understand the operation mechanism of NSTIS.So it will be an important research direction.For policy-making,identifying the time lags effects of NSTIS helps policy-makers better understand why an innovation input policy may have no significant effect on innovative performance growth immediately.Therefore,public policy should safeguard the persistence of innovative investment and avoid everything that could disturb the persistent transformation process.The assessment approach for national science,technology and innovation capabilities is expected to provide scientific and effective technical support for countries to make horizontal comparison and improve innovation policies.