Stage Analysis and Prediction of Green Finance via Sliding Window Decision Tree and Clustering

This study focuses on the core issue of objectively assessing and categorizing the development stages of green finance in Chinese provinces. Based on green finance-related data from 31 provincial-level administrative regions in mainland China from 1990 to 2021, K-means cluster analysis and decision tree modeling, combined with the sliding window mechanism, are used to scientifically categorize the development stages of green finance, while enhancing the accuracy of prediction in the time dimension. The results show that the development of green finance in China can be classified into four stages: germination, development, deepening and maturity, and the key jump factors between the stages include the proportion of green credit, green insurance and green bonds. These findings provide actionable insights for policymakers to prioritize interventions and optimize green financial policies tailored to regional characteristics. The code and dataset can be found at https://github.com/leidiya01/SWDTC.