DYNAMIC EVALUATION OF OPPORTUNITY INEQUALITY AND POLICY OPTIMIZATION VIA THE HST-GBM MODEL

Abstract

In the context of China’s rapid economic development and structural transformation, inequality of opportunity has become a major obstacle to upward mobility and the achievement of common prosperity. This study proposes an innovative Hybrid Spatiotemporal Gating-Boosted Model (HST-GBM), integrating LSTM, XGBoost, fixed-effects models, and gated recurrent units, to address the limitations of traditional methods in capturing nonlinear, heterogeneous, and intertemporal dynamics. Using CFPS panel data from 2010 to 2022, the study examines the evolution of opportunity inequality and its key drivers. First, the model effectively quantifies the relative impact of individual effort and environmental factors, revealing a long-term trend of declining inequality and increasing influence of personal effort. Second, the analysis shows that opportunity structures are highly responsive to policy changes, with effort-related impacts intensifying during reform windows. Finally, the study offers targeted policy recommendations that emphasize adaptive algorithms and effort-oriented approaches. Overall, this research contributes by advancing methodological innovation, enriching the theoretical understanding of dynamic inequality, and providing empirical support for policies aimed at promoting social equity and long-term stability.