Metody Ilościowe w Badaniach Ekonomicznych

Energy Demand as a Moderator of the Energy Transition in Africa: A Two-Way Fixed Effects Panel Analysis for 27 Countries, 2000–2022

Michal Thlon — 2026-09-30

Analysing a balanced panel of 27 African economies over 2000–2022, I document a “renewable share paradox”: in 63% of countries, absolute renewable capacity grew while its share in total installed capacity declined from 54.9% to 46.3%. Using a two-way fixed effects estimator, I find a robust negative within-country relationship between log GDP per capita and the renewable capacity share (β ranging from –10.9 to –21.7 pp), supporting the demand-squeeze hypothesis. Targeted public investment in renewables can partially offset this effect.

Short-Term WIG20 Volatility around Interest-Rate Decision Announcements by the Monetary Policy Council of the National Bank of Poland: Evidence from 2005–2025

Jakub Krawczyk — 2026-09-30

This study examines short-term WIG20 volatility around 235 interest-rate announcements by the Monetary Policy Council of the National Bank of Poland in 2005-2025. Volatility is measured from daily log returns in three- and five-day windows and analysed separately for decisions involving interest-rate changes and status quo decisions. Descriptive statistics show generally higher pre-announcement volatility and particularly pronounced skewness and kurtosis before interest-rate-change announcements. Wilcoxon signed-rank and Mann-Whitney U tests show that pre-announcement volatility is significantly higher only for decisions involving interest-rate changes in the three-day window. Between-group differences are also significant in the three-day window, but only at the 10% level.

Article 33a of the VAT Act - Improved Financial Liquidity for Importers in Poland and the Risk of Abuse

Jarosław Gorący — 2026-09-30

Article 33a of the Polish VAT Act is the Polish model of postponed VAT accounting, allowing import VAT to be settled in the VAT return instead of at customs clearance. Using aggregated data for 2020–2024, the article analyses the scale, commodity structure and control significance of this mechanism. The results show that it covered about one quarter of Poland’s imports and was concentrated mainly in industrial, raw-material and technological goods. It should therefore be viewed both as a liquidity tool and as an important area of customs and tax risk profiling.

Alfabetyzm statystyczny a pułapki myślenia. Analiza luki kompetencyjnej na styku edukacji średniej i wyższej

Ewa Wasilewska — 2026-09-30

Celem artykułu jest ocena poziomu alfabetyzmu statystycznego uczniów szkół średnich i studentów oraz analiza wybranych komponentów tej kompetencji. Badanie przeprowadzono wśród 432 uczniów i studentów z wykorzystaniem autorskiego testu sytuacyjnego. Wyniki wskazują na istnienie luki kompetencyjnej pomiędzy efektami formalnej edukacji matematyczno-statystycznej a poziomem alfabetyzmu statystycznego ujawnianym podczas interpretacji danych i rozumowania probabilistycznego. Analiza ujawniła anomalię edukacyjną: uczniowie realizujący rozszerzony program matematyki wykazali wyższy poziom alfabetyzmu statystycznego niż studenci po akademickim kursie statystyki.

Mitigating Class Imbalance in U.S. Financial Security: A Latent Feature Augmentation Approach for High-Purity Fraud Detection

Kaniz Sultana Chy — 2026-09-30

Credit card fraud detection is challenged by extreme class imbalance (< 0.2% fraud) and non-linear fraud patterns, causing traditional supervised models to suffer from reduced sensitivity and poor interpretability. This paper proposes a hybrid model combining a sparse Deep Autoencoder (15-node bottleneck) with Random Forest classification via feature augmentation—concatenating raw features with latent representations and reconstruction errors. Validation on 85,442 transactions yields PR-AUC of 0.6942, precision of 0.9412 ± 0.0318, and recall of 0.7968 ± 0.0354 across 5-fold cross-validation. While performance is statistically comparable to standard Random Forest (p > 0.05), the hybrid model offers enhanced interpretability: the Anomaly Score ranks as the second most important predictor, and t-SNE reveals distinct geometric separation of fraud clusters. These findings demonstrate that unsupervised feature augmentation provides a viable, interpretable solution for real-time fraud monitoring.