Replication data and code for “Petroleum balance and sectoral carbon responses to OPEC supply-news shocks”
Description
This dataset contains the submission-stage reproducibility materials for the manuscript “Petroleum balance and sectoral carbon responses to OPEC supply-news shocks.” The study combines country-by-sector fossil CO2 emissions, national petroleum consumption and production data, and externally identified OPEC supply-news shocks to examine whether pre-existing petroleum balance is associated with heterogeneous sectoral carbon responses under a common external petroleum-supply disturbance. The archive contains the derived analytical panels, locked sample definitions, estimation code, model outputs, dependent wild-bootstrap familywise inference, influence diagnostics, and temporal-support diagnostics underlying the manuscript results. Source data are derived from publicly available international databases as documented in the accompanying README and manuscript. No confidential or personally identifiable data are included. Primary source files are not redistributed in this archive. They remain externally available from the providers cited in the manuscript, including IEA-EDGAR/JRC emissions data, U.S. EIA international petroleum data, and the Känzig OPEC supply-news shock series. Accordingly, the archive reproduces the submitted analytical panels, model outputs, diagnostics, and analysis code, but is not presented as a self-contained raw-data-to-final-output package.
Files
Steps to reproduce
1. Download and extract the reproducibility archive. 2. Read README_ENERGY_POLICY_REPRODUCIBILITY.txt first. It describes the package structure, analytical scope, source-data boundary, locked sample definitions, and the relationship between the archived inputs and the submitted manuscript. 3. Open the B3 eligibility-audit archive to inspect the coefficient-blind sample eligibility rules, sample construction, locked country rosters, and coverage diagnostics. 4. Open the B4 expanded-sample response archive to reproduce the CORE68, BROAD72, and NEW21 analytical panels and the manuscript-facing point estimates. 5. Open the B5 dependent-wild-bootstrap archive to reproduce the joint four-horizon inference, simultaneous confidence intervals, calendar-time support diagnostics, and influence analyses. 6. The archived analytical inputs correspond to the country, sector, petroleum, and shock-series definitions documented in the manuscript and Supplementary Material. Primary source data remain available from the cited public data providers.
Institutions
- Kocaeli ÜniversitesiKocaeli, İzmit