A DTW-STIRPAT-Monte Carlo model for carbon emission prediction and dual-carbon pathway in Chongqing

Abstract

Under China’s dual-carbon goals, accurately predicting regional carbon emission trajectories and identifying key drivers are essential for differentiated mitigation policies. This study develops a hybrid DTW–STIRPAT–Monte Carlo model for Chongqing, a typical industrial city, using data from 2000 to 2022. Dynamic Time Warping (DTW) identifies key drivers; an extended STIRPAT model incorporating the Environmental Kuznets Curve and dual-carbon constraints is built, with ridge regression addressing multicollinearity; and Monte Carlo simulation quantifies future uncertainty. The results indicate that energy intensity and industrial value-added are the dominant drivers of carbon emissions in Chongqing, contributing 52.5% and 21.4%, respectively. Urbanization exhibits a negative elasticity with respect to emissions, reflecting the mitigation effects of compact urban development. Scenario-based projections suggest that under enhanced policy scenarios, Chongqing’s carbon emissions are likely to peak around 2030 and decline steadily toward 2060, despite increasing long-term uncertainty. The model demonstrates strong robustness to alternative distributional assumptions. Overall, the proposed DTW–STIRPAT–Monte Carlo framework shows strong generalization capability under limited data conditions and provides a transferable methodological tool for carbon peaking and neutrality pathway analysis in industrial cities.