Working Papers

The WNE Working Papers series has been published by the Faculty of Economic Sciences at the University of Warsaw since 2008.

The WNE Working Papers series provides a fast, open channel for disseminating research conducted at the Faculty of Economic Sciences, University of Warsaw. The papers hosted here are working versions (pre-prints) and may evolve as authors refine their analyses, incorporate feedback, or progress through formal peer-review. What you read is the current version released by the authors, timestamped and assigned DOI/ISSN identifiers to ensure precise citation and version tracking. Copyright remains with the authors, who may, at any time, upload a revised file or add a note directing readers to a later, peer-reviewed publication.

The Working Papers series accepts articles by research employees of the Faculty and publications from conferences organised at the Faculty of Economic Sciences at the University of Warsaw. Articles should be original research papers which have not been previously published, on the subject of economics.

Please send your paper by e-mail: dY-I{/SjN0?+^bv#C8%tEw|g@Qo]#[\Iz1ex7[+{{yRynt!EwlN]a]IB\ 

Please send 2 files:

(1) the main text without the title of the article and the authors (DOC/DOCX file) and

(2) the title page including: the title of the paper, the authors and their affiliation (DOC/DOCX file).

Please read the detailed editing requirements before submitting your text. 

Number of Results 50

WP(32/2026)526. Same Constraints in Changing Times? A Machine-Learning Approach to Childbearing and Fertility Intentions in Poland during and after the COVID-19 Pandemic

Authors: Kurowska Anna, Grabowska Magdalena, Świtała Maciej, Osiewalska Beata
While studies highlight COVID-19's disruptive impact on fertility, whether pandemic-era childbearing predictors remain informative post-crisis is less clear. Using Polish Familydemic panel data from 2021–2024 (N = 1,925, aged 18–49), …

WP526. Same Constraints in Changing Times? A Machine-Learning Approach to Childbearing and Fertility Intentions in Poland during and after the COVID-19 Pandemic

While studies highlight COVID-19's disruptive impact on fertility, whether pandemic-era childbearing predictors remain informative post-crisis is less clear. Using Polish Familydemic panel data from 2021–2024 (N = 1,925, aged 18–49), we apply theory-guided machine learning to evaluate a broad set of individual, employment, partnership, and family characteristics. Because short-term fertility intentions prove most predictive of childbearing, we model both actual births and firm positive intentions using random forest. The pandemic-trained model retained strong post-pandemic predictive performance (AUC-ROC = 0.86), demonstrating that pandemic-era drivers remain informative. Childbearing and intentions shared 16 of their top 20 predictors, highlighting common relational, demographic, and work–family factors. Relationship satisfaction, work–life balance, and domestic work division were particularly prominent, whereas prolonged school and childcare closures lowered predicted probabilities for both outcomes among parents. Ultimately, post-pandemic fertility dynamics reflect persistent constraints embedded in everyday family and working lives.


DOI: https://doi.org/10.33138/2957-0506.2026.32.526
Kurowska Anna, Grabowska Magdalena, Świtała Maciej, Osiewalska Beata
childbearing fertility intentions COVID-19 pandemic machine learning relationship satisfaction childcare closures

WP(31/2026)525. Co-adaptation: From MARL Non-stationarity to the Dynamics of Economic Adjustment

Authors: Lisicki Jan, Kopczewski Tomasz
Economic agents adapt their behavior to experience generated by interactions with other market participants. When several agents update their behavioral rules simultaneously, these adaptation processes become interdependent. We refer to this mechanis…

WP525. Co-adaptation: From MARL Non-stationarity to the Dynamics of Economic Adjustment

Economic agents adapt their behavior to experience generated by interactions with other market participants. When several agents update their behavioral rules simultaneously, these adaptation processes become interdependent. We refer to this mechanism 
as co-adaptation. The paper develops a formal representation that separates the continuation of an agent's own rule updates from cross-agent effects on subsequent updates. In a two-agent system, mutual influence can increase the cumulative effect of an initial change and prolong the resulting adjustment process. A simple model of price adjustment between two firms shows how these effects depend on adjustment speed and the response to a competitor's price change. With more agents, interdependencies between adaptation processes form a directed network. This perspective links economic adjustment to multi-agent reinforcement learning, where learning by one agent changes the environment experienced by others, and clarifies when such non-stationarity is part of economic dynamics rather than only a computational difficulty.


DOI: https://doi.org/10.33138/2957-0506.2026.31.525
Lisicki Jan, Kopczewski Tomasz
economic adjustment multi-agent reinforcement learning adaptive learning agent-based computational economics co-adaptation

WP(21/2026)515. Beyond Weeks Closed: Reassessing Pandemic Educational Disruption with TIMSS 2003–2023

Authors: Binkiewicz Marzena, Pokropek Artur
Using six TIMSS cycles (2003–2023), UNESCO school-closure data, and World Bank GDP indicators, we estimate deviation-from-trend and continuous-treatment Difference-in-Differences models for mathematics achievement, liking learning mathematics, …

WP515. Beyond Weeks Closed: Reassessing Pandemic Educational Disruption with TIMSS 2003–2023

Using six TIMSS cycles (2003–2023), UNESCO school-closure data, and World Bank GDP indicators, we estimate deviation-from-trend and continuous-treatment Difference-in-Differences models for mathematics achievement, liking learning mathematics, and confidence in mathematics. Mathematics achievement declined substantially relative to pre-pandemic trends, but closure duration did not consistently explain cross-national variation in achievement losses. Unexpectedly, longer closures were associated with slightly higher liking learning mathematics, particularly among Grade 4 students in higher-GDP countries, while confidence in mathematics remained largely unchanged. These findings suggest a decoupling of cognitive and affective outcomes during the pandemic and indicate that school-closure duration alone is an insufficient measure of educational harm. Comparative evaluations of pandemic education policies should therefore incorporate indicators of instructional continuity, remote-learning quality, and recovery efforts alongside formal school-closure measures.


DOI: https://doi.org/10.33138/2957-0506.2026.21.515
Binkiewicz Marzena, Pokropek Artur
Mathematics achievement non-cognitive factors distance learning school closures TIMSS math anxiety self-efficacy

WP(18/2026)512. Painting Price: A Machine Learning Approach to Art Valuation. Proof of Concept and Market Structure Diagnosis

Authors: Okhrimenko Kostiantyn
This paper investigates the feasibility of predicting art prices using machine learning methods applied to a dataset of 20,905 paintings and drawings scraped from the Artsper online marketplace. We test tree-based ensemble models (Decision Tree, Rand…

WP512. Painting Price: A Machine Learning Approach to Art Valuation. Proof of Concept and Market Structure Diagnosis

This paper investigates the feasibility of predicting art prices using machine learning methods applied to a dataset of 20,905 paintings and drawings scraped from the Artsper online marketplace. We test tree-based ensemble models (Decision Tree, Random Forest, XGboost) and deep learning architectures (MLP, CNN, Fusion) on both tabular metadata and hand-crafted image features. Results consistently show poor predictive performance across all model types and feature sets. We argue that this outcome is not a methodological failure but a substantive finding: it constitutes a diagnosis of the market structure of contemporary art. Drawing on hedonic pricing theory (Rosen 1974), the sociology of cultural fields (Bourdieu 1993), and the economics of valuation (Velthuis 2005; Beckert and Rössel 2013), we propose a three-layer model of art price determinants: physical attributes (observable and partially captured by models), visual-aesthetic features (observable but poorly quantifiable), and narrative-reputational capital (largely unobservable in cross-sectional platform data).


DOI: https://doi.org/10.33138/2957-0506.2026.18.512
Okhrimenko Kostiantyn
art market machine learning hedonic pricing art valuation cultural economics XGboost CNN

WP(14/2026)508. Know Thyself: A Methodological Manifesto for Teaching Microeconomics Through Epistemic Provocation

Authors: Kopczewski Tomasz
This paper documents and formalises the Know Thyself method, a teaching approach developed through more than thirty years of university teaching practice. Its starting diagnosis is that students often learn economic models without experiencing the as…

WP508. Know Thyself: A Methodological Manifesto for Teaching Microeconomics Through Epistemic Provocation

This paper documents and formalises the Know Thyself method, a teaching approach developed through more than thirty years of university teaching practice. Its starting diagnosis is that students often learn economic models without experiencing the assumptions that make those models necessary. The method reverses the usual sequence: experience before theory, data as a mirror before abstraction, and questions before answers. Its empirical core is not the experiment narrowly understood, but ad hoc research: classroom experiments, surveys, simulations, valuation tasks, and replication laboratories that make learners’ own assumptions visible. Four case studies — expected value, ergodicity, market equilibrium, and the rationality of altruism — illustrate how the method converts declarative knowledge into reflective practice. Artificial intelligence gives the method scale by lowering the cost of surveys, dashboards, simulations, and replication protocols. The paper’s practical conclusion is simple: change the order. Ask first. Teach later.


DOI: https://doi.org/10.33138/2957-0506.2026.14.508
Kopczewski Tomasz
economics education epistemic provocation experiential learning replication expected value ergodicity market equilibrium AI in education science curiosity

WP(7/2026)501. Causal Inference under Algorithmic Interference: Identification and Estimation without SUTVA in Platform Economies

Authors: Ryłow Jakub
The Stable Unit Treatment Value Assumption (SUTVA) fails systematically in platform economies where a deterministic algorithm mediates all interactions, making interference structural and mechanistically knowable. We introduce algorithmic interferenc…

WP501. Causal Inference under Algorithmic Interference: Identification and Estimation without SUTVA in Platform Economies

The Stable Unit Treatment Value Assumption (SUTVA) fails systematically in platform economies where a deterministic algorithm mediates all interactions, making interference structural and mechanistically knowable. We introduce algorithmic interference — a structural potential-outcomes model in which spillovers flow through the platform's known decision rule — and construct the Debiased Algorithmic Instrumental Variable (DAIV) estimator: a cross-fitted semiparametric procedure combining Double Machine Learning with the IV equation implied by the algorithmic mechanism. Under local algorithmic monotonicity (LAM), both the ATE and CATE are point identified; without LAM the sharp identified set is characterised. DAIV is sqrt(n)-consistent, asymptotically normal, and semiparametrically efficient, with a formal LAM test supplied. A synthetic ride-sharing example (n = 10,000) shows that standard DML overstates the treatment effect by 52% relative to DAIV; a Hausman-type specification test strongly rejects no algorithmic interference.


DOI: https://doi.org/10.33138/2957-0506.2026.7.501
Ryłow Jakub
algorithmic interference SUTVA potential outcomes causal identification double machine learning platform economies instrumental variables semiparametric efficiency partial identification

WP(30/2025)493. Is smooth Energiewende possible? Improving the performance of climate policies in Germany by optimizing the risk of electricity delivery

Authors: Bandurski Jakub, Hałatek Eliza, Łaziński Adam, Künstler Michał
The Energiewende is a deep-rooted notion in the German economy. The main goal is to achieve climate neutrality by transitioning to renewable energy sources. However, the feasibility of this transition is partially hindered by power grid congestion, w…

WP493. Is smooth Energiewende possible? Improving the performance of climate policies in Germany by optimizing the risk of electricity delivery

The Energiewende is a deep-rooted notion in the German economy. The main goal is to achieve climate neutrality by transitioning to renewable energy sources. However, the feasibility of this transition is partially hindered by power grid congestion, which undermines system efficiency and leads to both economic and environmental costs. We address this issue by making a prediction of the likelihood of congestion occurrence within the German TenneT DE electricity network in the years 2020-2023. We propose a twofold approach offering a combination of advanced econometric models and state-of-the-art machine learning methods. We offer separate solutions for up congestion when additional energy needs to be pushed to the network as well as down congestion when energy needs to be pulled away from the network. Analyzing the CatBoost with XAI, we identify factors that play a significant role in driving redispatch events within the German electricity network.


DOI: https://doi.org/10.33138/2957-0506.2025.30.493
Bandurski Jakub, Hałatek Eliza, Łaziński Adam, Künstler Michał
energiewende econometrics machine learning climate policy catboost

WP(29/2025)492. Childbearing in the Knowledge-Based Society: Job-Related Learning Demands and the Transition to Parenthood in Germany

Authors: Luo Chen, Jarosz Ewa, Matysiak Anna
This study investigates the relationship between learning demands at work and the transition to parenthood in Germany. As a consequence of technological progress and intensifying global competition, workplace learning is no longer an optional path to…

WP492. Childbearing in the Knowledge-Based Society: Job-Related Learning Demands and the Transition to Parenthood in Germany

This study investigates the relationship between learning demands at work and the transition to parenthood in Germany. As a consequence of technological progress and intensifying global competition, workplace learning is no longer an optional path to career advancement but has become an essential job demand. Consequently, it absorbs time and energy that could otherwise be devoted to family formation, prompting individuals to postpone childbearing or have fewer children. Yet, the fertility implications of this structural change have not been systematically examined. This study addresses this gap by analysing how job-related high learning demands relate to the transition to the first birth. The results indicate that individuals in jobs with high learning demands, both men and women, tend to delay the transition to the first birth. However, these delays do not appear to preclude them from becoming parents later, suggesting a postponement rather than a withdrawal from parenthood.


DOI: https://doi.org/10.33138/2957-0506.2025.29.492
Luo Chen, Jarosz Ewa, Matysiak Anna
fertility childbearing learning demands labour market transformation work-family conflict

WP(25/2025)488. Application of Deep Reinforcement Learning to At-the-Money S&P 500 Options Hedging

Authors: Bracha Zofia, Michańków Jakub, Sakowski Paweł
This paper explores the application of deep Q-learning to hedging at-the-money options on the S&P 500 index. We develop an agent based on the Twin Delayed Deep Deterministic Policy Gradient (TD3) algorithm, trained to simulate hedging decisions w…

WP488. Application of Deep Reinforcement Learning to At-the-Money S&P 500 Options Hedging

This paper explores the application of deep Q-learning to hedging at-the-money options on the S&P 500 index. We develop an agent based on the Twin Delayed Deep Deterministic Policy Gradient (TD3) algorithm, trained to simulate hedging decisions without making explicit model assumptions on price dynamics. The agent was trained on historical intraday prices of S&P 500 call options across years 2004 to 2024, using a single time series of six predictor variables: option price, underlying asset price, moneyness, time to maturity, realized volatility, and current hedge position. A walk-forward procedure was applied for training, which lead to nearly 17 years of out-of-sample evaluation. The performance of the deep reinforcement learning (DRL) agent is benchmarked against the Black–Scholes delta hedging strategy over the same time period. We assess both approaches using metrics such as annualized return, volatility, information ratio, and Sharpe ratio. To test models’ adaptability, we performed simulations across varying market conditions and added constraints such as transaction costs and risk-awareness penalties. Our results show that the DRL agent can outperform traditional hedging methods, particularly in volatile or high-cost environments, highlighting its robustness and flexibility in practical trading contexts. While the agent consistently outperforms delta hedging, its performance deteriorates when the risk-awareness parameter is higher. We also observed that the longer the time interval used for volatility estimation, the more stable the results.


DOI: https://doi.org/10.33138/2957-0506.2025.25.488
Bracha Zofia, Michańków Jakub, Sakowski Paweł
Deep learning Reinforcement learning Double Deep Q-netwoorks options market options hedging deep hedging

WP(22/2025)485. A survey of statistical arbitrage pair trading with machine learning, deep learning, and reinforcement learning methods

Authors: Sun Yufei
Pair trading remains a cornerstone strategy in quantitative finance, having consistently attracted scholarly attention from both economists and computer scientists. Over recent decades, research has expanded beyond traditional linear frameworks&mdash…

WP485. A survey of statistical arbitrage pair trading with machine learning, deep learning, and reinforcement learning methods

Pair trading remains a cornerstone strategy in quantitative finance, having consistently attracted scholarly attention from both economists and computer scientists. Over recent decades, research has expanded beyond traditional linear frameworks—such as regression- and cointegration-based models—to embrace advanced methodologies, including machine learning (ML), deep learning (DL), reinforcement learning (RL), and deep reinforcement learning (DRL). These techniques have demonstrated superior capacity to capture nonlinear dependencies and complex dynamics in financial data, thereby enhancing predictive performance and strategy design.
Building on these academic developments, practitioners are increasingly deploying DL models to forecast asset price movements and volatility in equity and foreign exchange markets, leveraging the advantages of artificial intelligence (AI) for trading. In parallel, DRL has gained prominence in algorithmic trading, where agents can autonomously learn optimal trading policies by interacting with market environments, enabling systems that move beyond price prediction to dynamic signal generation and portfolio allocation.
This paper provides a comprehensive survey of ML-, DL-, RL-, and DRL-based approaches to pair trading within quantitative finance. By systematically reviewing existing studies and highlighting their methodological contributions, it offers researchers a structured foundation for replication and further development. In addition, the paper outlines promising avenues for future research that extend the application of AI-driven methods in statistical arbitrage and market microstructure analysis.


DOI: https://doi.org/10.33138/2957-0506.2025.22.485
Sun Yufei
Pair Trading Machine Learning Deep Learning Reinforcement Learning Deep Reinforcement Learning Artificial Intelligence Quantitative Trading