Correlation matrix of main variables (n = 19956).

This study investigates the impact of data assets on enterprise persistent innovation using panel data from Chinese A-share listed firms from 2011 to 2022. The results indicate that data assets significantly enhance both the inputs and outputs of enterprise persistent innovation, with the findings remaining robust under endogeneity tests. Mediation analysis reveals that data assets influence enterprise persistent innovation through three key channels: process innovation, business innovation, and technological innovation. The development of digital finance positively moderates this relationship across three dimensions of coverage, depth, and digitalization, indicating that digital finance amplifies the persistent innovation value of data assets. Heterogeneity analyses reveal that the persistent innovation input improves more in non-state-owned enterprises, digitally advanced firms, and non-manufacturing sectors, whereas output enhancement is most evident in large enterprises, highly digitalized firms, and organizations with strong absorptive capacity. These findings contribute to a deeper understanding of data-driven persistent innovation and provide valuable insights for policymakers developing data markets, and for firms formulating data strategies aligned with their capabilities.本研究基于2011至2022年中国A股上市公司的面板数据（panel data），探究数据资产（data assets）对企业持续创新的影响。研究结果表明，数据资产可显著提升企业持续创新的投入与产出水平，且该结论在内生性检验下依然稳健。中介效应分析显示，数据资产通过流程创新、商业模式创新与技术创新三大核心路径影响企业持续创新。数字金融（digital finance）的发展在覆盖广度、使用深度与数字化程度三个维度上对该关系产生正向调节作用，即数字金融会放大数据资产的持续创新价值。异质性分析结果表明，非国有企业、数字化程度较高的企业以及非制造业的持续创新投入提升更为显著，而持续创新产出的提升则在大型企业、高数字化企业与吸收能力较强的组织中最为明显。本研究的发现有助于深化对数据驱动型持续创新的理解，并为政策制定者完善数据市场建设、企业制定契合自身能力的数据战略提供了极具价值的参考启示。