Patent Quality, Institutions, and Innovation Leadership in FinTech: A Multidimensional Machine Learning Analysis
构建了一个包含技术范围、法律稳健性和市场响应性的专利质量指数（PQI），基于2000-2023年全球96,657件金融科技专利，发现专利质量存在显著异质性，且受监管质量、腐败控制和外商直接投资等制度因素影响。
Firms in technology-intensive industries require reliable tools to evaluate patent quality and guide R&D resource allocation. Conventional indicators such as patent counts or citation totals provide incomplete signals and often privilege volume over strategic substance. This study develops and applies a Patent Quality Index (PQI), a multidimensional framework integrating technological scope, legal robustness, and market responsiveness to capture the intrinsic quality of patents. Using a global dataset of 96,657 FinTech patents filed between 2000 and 2023, the PQI is constructed through complementary statistical and machine-learning methods and applied to compare patent quality across firms and jurisdictions. Results reveal substantial heterogeneity in patent quality, clear trade-offs between portfolio size and average quality at the firm level, and systematic cross-country differences. Specifically, regulatory quality, corruption control, and foreign direct investment significantly shape national patterns of patent quality, linking firm-level innovation strategies with broader governance environments. By providing a scalable and theory-grounded measure of patent quality, the PQI advances innovation measurement and offers practical guidance for managers, policymakers, and investors seeking to benchmark portfolios, prioritise high-value patents, and design policies that promote substantive innovation. Although demonstrated in the FinTech sector, the framework is adaptable to other technology-driven industries.