| 中文题名： | 研发经费对专利产出的滞后效应研究——基于美国上市公司面板数据的分析 |
| 姓名： | |
| 保密级别： | 公开 |
| 论文语种： | 中文 |
| 学科代码： | 120102 |
| 学科专业： | |
| 学生类型： | 学士 |
| 学位： | 管理学学士 |
| 学位年度： | 2021 |
| 学校： | 北京师范大学 |
| 校区： | |
| 学院： | |
| 第一导师姓名： | |
| 第一导师单位： | |
| 提交日期： | 2021-06-19 |
| 答辩日期： | 2021-05-12 |
| 外文题名： | Research on the Lag Effect of R&D Expenditure on Patent Output——Analysis based on panel data of U.S. listed companies |
| 中文关键词： | |
| 中文摘要： |
目前研发经费与专利产出的相关研究大多是在不划分行业的情况下，对所有公司进行了实证研究，而行业层面的研究较少。因此，本研究在行业层面上分析研发经费对专利产出的滞后效应，比较行业之间的区别，了解不同行业的研发特征以及行业自身的独特性，为公司的研发投入策略起到借鉴作用。本研究基于243家美国研发顶尖公司从2012年到2018年的研究经费与专利授权量面板数据，进行基本特征分析和相关性分析，最后利用多项式分布滞后模型分析每个行业的研发经费对专利产出的滞后影响结构。结果发现，大部分行业的研发经费与专利授权量显著正相关，但也有一些行业的两个变量之间没有显著的相关关系。其次，不同行业的滞后分布结构各不相同，有些行业的研发经费对专利授权量的滞后效应随时间先增后减，有些行业则相反；不同行业的滞后长度也不同，但都在3-4年范围内。最后，研究发现在不划分行业时，所有243家公司的相关性系数小于任何一个划分行业时的相关性系数；所有243家公司分布滞后模型的滞后长度是1年，这与其他行业3-4年的滞后长度有所差异。因此本研究在一定程度上证明了研究研发经费对专利产出的滞后效应时，需要将企业按照行业进行划分，在行业层面上进行滞后效应研究。
﹀
|
| 外文摘要： |
At present, most of the researches related to R&D expenditure and patent output have conducted empirical studies on all companies without industry division, while few studies have been conducted at the industry level. Therefore, this study analyzes the lag effect of R&D expenditure on patent output at the industry level, compares the differences between industries, and understands the R&D characteristics of different industries and the uniqueness of the industries themselves, so as to play a reference role for the company's R&D investment strategy. Based on the panel data of R&D expenditure and patent grant of 243 top US R&D companies from 2012 to 2018, this study conducted basic characteristic analysis and correlation analysis. Finally, the polynomial distributed lag model was used to analyze the lag impact structure of R&D expenditure on patent output in each industry. The results show that most industries have a significant positive correlation between R&D expenditure and the number of patents granted, but some industries have no significant correlation between the two variables. Secondly, the lag distribution structure of different industries is different. In some industries, the lag effect of R&D expenditure on the number of patent grants increases first and then decreases with time, while in some industries, it is the opposite. The lag length varies from industry to industry, but is in the range of 3-4 years. Finally, it is found that the correlation coefficients of all 243 firms are smaller than those of any industry. The lag length of the distributed lag model for all 243 companies is 1 year, which is different from the 3-4 year lag length for other industries. Therefore, to a certain extent, this study proves the lag effect of R&D expenditure on patent output, and it is necessary to divide enterprises according to their industries and conduct research on the lagging effect at the industry level.
﹀
|
| 参考文献总数： | 41 |
| 插图总数： | 5 |
| 插表总数： | 7 |
| 馆藏号： | 本120102/21005 |
| 开放日期： | 2022-06-19 |
谷歌
