Path Dependency in Institutional Change and Port System Development: A Spatiotemporal Dynamic Analysis of the Greater Bay Area Based on Data Science

This study aims to explore how institutional changes impact port system development and reveal the path dependency characteristics within. Focusing on the Guangdong-Hong Kong-Macao Greater Bay Area, the project applies data science methods and incorporates key indicators such as Concentration Ratio (CR), Herfindahl-Hirschman Index (HHI), and Dynamic Shift-Share Analysis (DSSA) to quantify the influence of institutions on port systems. Additionally, GIS technology is used for spatiotemporal dynamic analysis, displaying the spatial evolution of the port system under various institutional
contexts. Machine learning models, including ARIMA and LSTM, are introduced to predict future port development trends, providing scientific support for regional economic decision-making. This study not only fills gaps in quantitative analysis of institutional impacts on ports domestically and internationally but also offers actionable policy recommendations for port planning and policy formulation. Ultimately, through empirical research and quantitative analysis, the project reveals the complex relationship between institutional changes and port system development, offering decision-making support for port managers and policymakers in an evolving global environment.