Evaluating the Science and Technology Finance Ecosystem in the Guangdong-Hong Kong-Macao Greater Bay Area based on XGBoost Algorithm

This study examines the evolution of the science and technology financial ecosystem in the Guangdong-Hong Kong-Macao Greater Bay Area from 2015 to 2023. A three-dimensional evaluation indicator system tailored to the region's unique characteristics was developed. Using entropy weighting, coupling coordination, and symbiosis models, the maturity and coordination of the ecosystem across 11 cities was evaluated. Results indicate that Hong Kong, Shenzhen, and Guangzhou are leaders in science and technology finance development. While core cities significantly contribute to this sector, the coupling degree among cities is improving, with Shenzhen achieving a deep coupling phase. This study recommends collaboration among cities to enhance the regional ecosystem, establish a multi-level direct financing market, and foster a highly open and cooperative environment for science and technology finance. To forecast R&D Spending and high-tech industry contribution, this paper utilizes the XGBoost Machine Learning (ML) algorithm. Experiments were conducted using the Guangdong region data. A detailed experiment analysis shows that the proposed algorithm outperforms well in terms of MSE, MAE, and R2 when compared to the other algorithms.