Abstract
In this paper, we propose a new, entirely data-driven and time-consistent index of environmental performance, the Data-Driven Environmental Performance Index (DDEPI), computed annually for 168 countries from 2015 to 2022. While traditional indexes often rely on expert-based weights, the DDEPI is built on empirical data using Robust Principal Component Analysis (Robust PCA) to identify variables that account for the largest share of variance. To ensure inter-temporal comparability, the methodology incorporates an explicit alignment step that matches yearly components to a fixed reference basis, addressing the limitations of traditional PCA in longitudinal settings. The index is primarily driven by indicators related to air quality, drinking water, and biodiversity. We validate the DDEPI by analyzing external socio-economic groupings and by conducting a diagnostic screening of country-specific structural breaks and volatility. Our results highlight the critical role of institutional quality and natural resource management in achieving sustainability goals, and provide a robust framework for monitoring environmental trajectories over time. •Proposes a time-consistent Data-Driven Environmental Performance Index (DDEPI).•Applies Robust PCA on pooled-standardized yearly slices for 168 countries (2015–2022).•Aligns yearly PCA components to a pooled reference to ensure inter-temporal comparability.•Aggregates aligned scores with fixed variance weights; includes an OECD-style robustness check.•Validates the index via stability and reconstruction diagnostics and interpretable group dynamics.