Forecasting Missing Persons Trajectories Using Autoregressive Models: A Jamaican Case Study

Abstract

This study analyses the temporal dynamics of missing persons, returned individuals, and unresolved cases in Jamaica using an autoregressive modelling framework. Annual data from 2018 to 2024 were examined to estimate persistence patterns and forecast trends for 2025-2030. A first-order autoregressive (AR(1)) specification was applied to model structural dependence and long-run equilibrium behaviour in each series. The current findings indicate strong temporal persistence across all variables, with autoregressive coefficients exceeding 0.90, suggesting slow adjustment to shocks and high structural inertia. The forecast projections indicate gradual convergence toward long-run equilibrium levels under stable institutional conditions. The results highlight the importance of sustained recovery capacity and policy intervention to reduce the stock of unresolved cases. The study contributes to the limited quantitative literature on the dynamics of missing persons in Jamaica and demonstrates the utility of time-series modelling for policy planning and institutional monitoring.