Efficiency evaluation of China’s high-tech industry with a dynamic network data envelopment analysis game cross-efficiency model

When using Data Envelopment Analysis (DEA) to evaluate the efficiency of a high-tech industrial system, it is necessary to consider the operating process in each period and the dynamic interdependence between periods of the system. Meanwhile, the shortcomings of the DEA self-evaluation mode cannot be ignored. However, few studies deal with these three problems in a unified framework. Therefore, this paper improves the DEA game cross-efficiency model to the dynamic network DEA game cross-efficiency model to evaluate the efficiencies of high-tech industries in 27 provincial-level regions of China from 2011 to 2015. The main evaluation results are as follows. Regarding overall efficiency, China’s high-tech industry still has approximately 45% room for improvement, and the development of adjacent regions is unbalanced. There are 18 regions with low Research and Development (R&D) efficiencies and 8 with low commercialization efficiencies. From a national perspective, R&D efficiency displays an inverted U-shaped trend, commercialization efficiency shows a U-shaped trend, and overall efficiency increases slightly during the study period. In addition, R&D efficiency has a greater impact on overall efficiency than commercialization efficiency does. The reasons are analyzed, and recommendations are provided based on the evaluation results to improve the efficiency of China’s high-tech industry.