Evaluation on regional innovation performance of China's patent data based on two-stage model

Being paid great attention to its operating performance, patent is the key element for a country or area to gain long-term advantage in various competitions. This paper, based on Chinese patent data, proposes a two-stage model with factor analysis and clustering analysis. It also proposes a patent innovation performance evaluation system in view of the input-output index of twenty patents, giving factor analysis to principal components, combining the various indexes to create new factors through calculating, extracting and aggregating, and then with this new factor score, this article evaluates the patent innovation performance systematically using the clustering analysis. The results show that, obviously, patent modification can promote innovation patent performance, especially the second modification. And at last, this article provides suggestion and development strategies to the improvement of patent system.