Evaluation And Analysis Of R&D Efficiency Of China's High-tech Industry Based On Common Weight DEA

Recent years,the state attaches great importance to the development of R&D activities,promoting R&D cooperation through financial appropriations and policy guidance,so as to achieve industrial technology innovation and the upgrading of industrial structure.The R&D inputs(R&D full-time personnel equivalent,R&D expenditure,new product development expenditure,etc.)and R&D outputs(new product sales revenue,new product exports,etc.)of high-tech industry all account for more than 27% of the national level,which occupies an important strategic position in China's R&D activities.Data Envelopment Analysis(DEA)is a data-driven tool for performance evaluation that is well suited for evaluating the relative efficiency of multi-input and multi-output decision-making units(DMUs),and is also the most commonly used method for R&D efficiency evaluation.Many scholars have used DEA to evaluate the R&D efficiency of China's high-tech industry.Undesirable outputs occur during R&D activities.The existing approaches that deal with undesirable outputs tend to either increase the efficiency scores of DMUs or keep the efficiency scores constant and do not allow undesirable outputs to achieve the opposite effect on the efficiency scores,which is inconsistent with the characteristics of undesirable outputs.To solve this problem,this paper constructs an aggregate DEA method based on common weights and undesirable outputs,calculates the aggregate weights of undesirable outputs through the common weight model,and integrates the undesirable output data into the desirable output data to obtain modified desirable output data.Taking these modified data as the final outputs,this paper calculates the R&D efficiency scores of high-tech industry of 30 China's regions in 2014,compares the results with the results without considering the undesirable outputs and the results of several existing methods,and illustrates the impact of undesirable outputs on the R&D efficiency of China's high-tech industry.The inputs and outputs of R&D activities are lagging behind in time.Considering time lag effects,researchers have two main processing approaches when evaluating the performance of R&D activities with DEA.One is to select input and output data for different periods,that is,to use the outputs of the latter period or the latter two periods to match the inputs of the current period;the other is to use the inputs of the current period and the previous period or the previous two periods to produce the outputs.The former is quite different from the actual R&D situation,and the latter will cause the reuse of the input data.This paper considers the time lag of R&D activities as two years,and introduces parameter ? to completely match the current inputs to the outputs of the current and the latter two periods.Assuming that there are a group of virtual DMUs with three processes inside,this paper calculates the values of the set of parameters by using a new common weight DEA model with the objective of the maximum overall efficiency.On this basis,the inputs are fully and non-repetitively distributed to various outputs,and the efficiency scores of R&D activities in 30 regions of China from 2014 to 2016 are calculated.The results of the new method are compared with the method without considering time lag effects and the method of discontinuous data and analyzed to illustrate the impact of time lag effects on the R&D efficiency of China's high-tech industry.