<![CDATA[Poverty is a multidimensional issue influenced by the interaction of various economic and social indicators. However, most previous studies have relied on static regression approaches, which fail to capture dynamic feedback mechanisms over time. This study analyzes the dynamic interactions among poverty headcount, population, the Open Unemployment Rate (OUR), real Gross Regional Domestic Product (GRDP), and the Human Development Index (HDI) in Aceh Province, Indonesia. Utilizing official provincial-level annual time-series data from Statistics Indonesia (BPS) for 2005–2024 (n=20), this study employs a Vector Autoregression (VAR) framework, including the Engle–Granger cointegration test, Error Correction Model (ECM), Impulse Response Function (IRF), and Variance Decomposition (VD). To preserve degrees of freedom and avoid over-parameterization given the short sample size, optimal lag lengths were strictly constrained based on information criteria. The findings indicate no significant short-run causal relationships among variables, except from poverty to HDI. The ECM estimation reveals that population, OUR, GRDP, and HDI do not significantly affect poverty in the short run, but exert strong cointegrating effects in the long run. IRF shows that shocks gradually return to equilibrium, while VD identifies GRDP and poverty as the most dominant shock drivers in the system. These findings demonstrate that poverty alleviation in Aceh requires long-term structural policies focused on inclusive economic growth, human capital enhancement, and labor market expansion rather than short-term interventions.]]>
