Abstract:     It is very important to study the factors that influence the examination period of high-profit patents to improve the efficiency of patent examination. This paper takes patent applications to the China National Intellectual Property Administration (CNIPA) from 1993 to 1999, which have lasted for 20 years as a sample. By observing the mechanism of patent examination by the Chinese Patent Office, the examination period could be divided into two stages: the preliminary examination period and the substantive examination period. The Violin Plot analysis and Cox regression model are used to conduct a phased study on influence factors on the examination period of high-profit patents. Specifically, the Violin Plot is used to analyze the distribution of application years, technical fields and applicant countries on the examination period in different stages. The Cox regression model is used to explore the correlation between the variables and the substantive examination period. The variables include the number of IPC classifications, the number of simple patent families, the number of priorities, the number of claims, the number of inventors, the number of references to patent literature/non-patent literature and the number of forward citations.According to the response of the Violin Plot to the data, the main findings are as follows: Firstly, the preliminary examination period and the substantive examination period are very unevenly distributed, with obvious discrete values. Secondly, the substantive examination period as a whole is longer than the preliminary examination period. Thirdly, the preliminary examination period is not greatly affected by the different application years and technical fields, but is significantly affected by the different applicant countries. Fourthly, the substantive examination period is significantly affected by the different application years, technical fields, and applicant countries. Judging from the final operating results of the Cox regression model, the substantive examination period of high-profit patents has a significant correlation with the number of IPC classifications, the number of priorities, the number of claims, the number of inventors and the number of forward citations, in which the number of IPC classifications, the number of priorities and the number of claims are the protective factors of the substantive examination period, while the number of inventors and the number of forward citations are risk factors. This means that the increase in the number of IPC classifications, the number of priorities and the number of claims will lead to longer substantive examination time, but the increase in the number of inventors and the number of forward citations will lead to shorter substantive examination time.In order to reasonably and effectively improve the examination efficiency and further promote the construction of a world-class intellectual property examination agency, based on the research conclusions of Violin Plot and Cox regression analysis, the following specific recommendations are put forward. In the preliminary examination stage, the first is to appropriately shorten the time limit for applicants to submit substantive examination. The second is to strictly control the reasons for the applicant to postpone the request for substantive examination. The third is to explore a demand-oriented examination model for applicants. In the substantive examination stage, the first is to strictly control the reasons for delaying the reply to the notice of examination opinion. The second is the precise positioning and fine management of the substantive examination departments in different technical fields. The third is to strengthen the examiner′s theoretical study of the corresponding technical field and update professional technical knowledge in time. The fourth is to require applicants to accurately locate the technical field of the patent application and clarify the subject of the patent. The fifth is to require applicants to reasonably limit the boundaries and quantity of claims. The sixth is to strengthen the construction of an intelligent retrieval system to improve the informatization level of examiners′ technical retrieval. When compressing the examination period, the quality of patent examination should also be guaranteed. In this regard, the recommendations are as follows. The first is to increase the outsourcing of existing technology searches. The second is that innovative examination modes such as separation of examination and search can be adopted. The third is to seek more international cooperation in the existing technology search and improve the existing technology database. The scope of the existing technology search could be significantly increased by using external knowledge. Through these, it is expected to achieve a balance between quality and efficiency of examination. 

Key words: high-profit patent, examination period, influence factor, Violin Plot, Cox regression, examination efficiency