Dynamic Panel Data Estimation of the Impact of Digital Finance on Technological Innovation Performance of Chinese Enterprises Based on BP Neural Network Analysis

Authors

  • Huimin Zhang School of Management, Shanghai University, Shanghai 200444, China

Abstract

Given the rapidly evolving field of digital finance, this study used a combination of back propagation (BP) network analysis and dynamic panel data estimation to investigate the impact of digital finance on the performance of Chinese enterprises in regard to scientific and technological innovation. The data was sourced from two open datasets: Quandl and Crunchbase. After cleaning and standardizing the data, a dynamic panel data model was constructed and analyzed using a BP neural network (BPNN).Through an evaluation of the BPNN model’s dynamic adaptability index, characteristic importance consistency index, and innovation contribution index, this study determined the relationship between digital finance and enterprise’s scientific and technical innovation performance. The results showed that for the two datasets, the dynamic adaptability indexes of BPNN model were above 82% and 83% respectively, indicating that the model has strong adaptability to data changes in different time periods. The consistency index of feature importance was above 86%, indicating the stability of the model when judging the importance of different input features. The findings demonstrate that the BPNN model has clear advantages when analyzing the relationship between digital finance and the scientific and technological innovation performance of enterprises. It can accurately capture the complex relationship between them and provide a reference for enterprises and policy makers. However, there are several limitations such as limited data set selection and incomplete evaluation indicators. Overall, the innovation contribution index is above 81% and 86%, which highlights the unique contribution of digital finance-related characteristics to the performance prediction of scientific and technological innovation of enterprises. By means of innovative evaluation indicators, the study also comprehensively analyzes the effect of applying the BPNN model in this field, thereby providing a valuable reference for subsequent research and practice.

Keywords: digital finance; scientific and technological innovation; back propagation network; dynamic adaptability; consistency of feature importance.

Cite As

H. Zhang, "Dynamic Panel Data Estimation of the Impact of Digital Finance on Technological Innovation Performance of Chinese Enterprises Based on BP Neural Network Analysis", Engineering Intelligent Systems, vol. 33 no. 5, pp. 535-544, 2025.



Downloads

Published

2025-09-01

Issue

Vol. 33 No. 5 (2025): International Journal of Engineering Intelligent Systems

Section

Article

License

The submission of a paper implies that, if accepted for publication, it will not be published elsewhere in the same form, in any language, without the prior consent of the publisher. Before publication, authors are requested to assign copyright to CRL Publishing Ltd. This allows CRL to sanction photocopying, and to authorize the reprinting of issues or volumes according to demand. Authors' traditional rights will not be jeopardized by assigning Copyright in this way, as they retain the right to reuse the material following publication, and to veto third-party publication.