A newspaper-based measure of economic policy uncertainty in Poland, covering January 2000 to June 2026. A continuous monthly index tracking uncertainty about the direction of economic policy in Poland—constructed from newspaper coverage and designed for empirical research.
Overview
EPU Index PL transforms 26 years of Polish newspaper coverage into a daily-constructed, monthly-published series for charting, comparison, and modelling. It detects policy uncertainty, identifies its level relative to a historical baseline, and reports a documented error rate.
The index is designed to make policy-related uncertainty in Poland measurable over time, interpretable against a fixed historical reference, and usable in empirical research.
Identifies newspaper articles that jointly refer to the economy, policy, and uncertainty using a documented keyword dictionary.
Places current readings against a fixed historical baseline, so periods of elevated or subdued uncertainty can be interpreted consistently.
Provides a reproducible time series with documented construction choices, validation checks, and an estimated classification error rate.
Methodology
A transparent pipeline connects article selection, newspaper coverage, time-series construction, historical splicing, and validation.
Economy ∧ Policy ∧ Uncertainty. The keyword approach identifies articles containing terms from all three conceptual groups. The dictionary has evolved through documented versions, with v5.1 representing the current version described in the methodology.
The index draws on a panel of six Polish newspaper titles, with title-level founding dates, coverage periods, and available publication days documented in the methodology.
The article counts are converted into a comparable aggregate series through five core operations:
Each step feeds the next, ending at mean = 100.
Standardisation uses the fixed 2000–2024 reference period (300 months), rather than a moving baseline.
Backward chain-splicing is used to connect the historical series across changes in newspaper availability and panel composition.
Validation combines several complementary checks rather than relying on one measure alone.
Articles in the human audit
Audit precision
Precision reported in the human validation audit.
Documented error rate
Estimated error rate from the same audit.
Cohen’s kappa (κ)
The index measures policy uncertainty as reflected in the selected newspaper coverage and operationalised through the documented dictionary and construction procedure. It should therefore be interpreted as a media-based indicator, with its source panel, keyword definition, historical splicing, and validation results considered when using it in empirical analysis.
Data
Explore the published monthly EPU Index PL series from January 2000 to June 2026. The index uses a fixed 2000–2024 baseline, normalised to an average of 100. Daily index readings and a machine-readable feed are available on request.
The embedded chart currently opens from July 2022. The published monthly series covers January 2000 to June 2026. The full historical range and accompanying revision notes are described in the project documentation.
Monthly observations are published for charting, comparison, and modelling.
Published series coverage
Baseline mean · 2000–2024
Baseline standard deviation
Monthly data licence
Citation
If you use the EPU Index PL in your research, please cite the accompanying paper.
@misc{landowska2026epupoland,
title={Economic Policy Uncertainty Index for Poland:
Data, Construction, and Validation},
author={Landowska, Alina M. and Kłopotek, Robert A.
and Zhang, He},
year={2026},
eprint={2611.05640},
archivePrefix={arXiv}
}