Szczepan Urjasz
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Szczepan Urjasz has a total of 9 co-authors. Here are the top co-authors by number of shared publications:
- Renata Karkowska
Department of Capital Markets and Investment, Faculty of Management, University of Warsaw, Warsaw, Poland
7 shared publications University of Warsaw
2 shared publicationsAlessia Palma
1 shared publication- Andrea Paltrinieri
Università Cattolica del Sacro Cuore
1 shared publication - Anna Vusata
Uniwersytet Warszawski
1 shared publication Florin Aliu
1 shared publication
Latest Publications
- Open Access
- 6 October 2026
- Published by Elsevier in SSRN Electronic Journal
How much tail spillover is real? Centring the residuals in quantile connectedness
Quantile connectedness measures spillovers in the tails of financial return distributions, and its characteristic finding is a U-shape: transmission far higher at extreme quantiles than at the median. This paper shows that the shape is largely an artefact of the reference implementation. The R package ConnectednessApproach supplies the generalised forecast error variance decomposition with the uncentred second moment of the quantile regression residuals in place of their covariance matrix. The two coincide under least squares but differ under quantile regression by the outer product of the residual means, a rank-one term the decomposition reads as dependence. The bias is exact and does not shrink as the sample grows. On independent series the index reads 82.8 in the tail against 0.4 at the median, and reads the same on series carrying genuine copula tail dependence. Centring the residuals removes the tail amplification on all eighteen panels the package distributes; across 158 specification cells the median residue is 2.2 per cent. Centring suffices wherever the quantile VAR is identified: scaling the rolling window to each system's parameter count takes the median residue from 11.6 to 1.3 per cent. Identification and time resolution then conflict, because an identified tail decomposition needs 200(k+1) observations, a decade of daily data for twelve series. On the eight bundled energy and commodity systems the corrected profile is flat on six, and the largest tail transmitter changes on half. The specification is correct; the implementation is not.
- 6 October 2026
- Published by Elsevier in SSRN Electronic Journal
Open AccessHow much tail spillover is real? Centring the residuals in quantile connectedness
Quantile connectedness measures spillovers in the tails of financial return distributions, and its characteristic finding is a U-shape: transmission far higher at extreme quantiles than at the median. This paper shows that the shape is largely an artefact of the reference implementation. The R package ConnectednessApproach supplies the generalised forecast error variance decomposition with the uncentred second moment of the quantile regression residuals in place of their covariance matrix. The two coincide under least squares but differ under quantile regression by the outer product of the residual means, a rank-one term the decomposition reads as dependence. The bias is exact and does not shrink as the sample grows. On independent series the index reads 82.8 in the tail against 0.4 at the median, and reads the same on series carrying genuine copula tail dependence. Centring the residuals removes the tail amplification on all eighteen panels the package distributes; across 158 specification cells the median residue is 2.2 per cent. Centring suffices wherever the quantile VAR is identified: scaling the rolling window to each system's parameter count takes the median residue from 11.6 to 1.3 per cent. Identification and time resolution then conflict, because an identified tail decomposition needs 200(k+1) observations, a decade of daily data for twelve series. On the eight bundled energy and commodity systems the corrected profile is flat on six, and the largest tail transmitter changes on half. The specification is correct; the implementation is not.
- Open Access
- 27 July 2026
- Published by Springer Nature in Financial Innovation
- Vol. 12 (1) , 141
Biodiversity and energy markets: How do spillover effects across stock indices differ between the US and Europe?
- 27 July 2026
- Published by Springer Nature in Financial Innovation
- Vol. 12 (1) , 141
Open AccessBiodiversity and energy markets: How do spillover effects across stock indices differ between the US and Europe?
- Open Access
- 28 April 2026
- Published by Elsevier in SSRN Electronic Journal
Banking, Insurance Sector and Energy Transition: Volatility Spillovers Between U.S. and European Markets
This paper investigates the volatility spillovers between financial sector Exchange Trade Funds (ETFs) and energy markets, including fossil-based and clean energy assets, across U.S. and European markets. Using a TVP-VAR model, we quantify dynamic connectedness and assess systemic risk transmission, especially during periods of heightened uncertainty. Results show strong within-group connectedness in both financial and oil sectors, while clean energy emerges as a growing transmitter of shocks, particularly in the U.S. market. The analysis also highlights the distinct roles of financial institutions in volatility propagation, with European banks and insurers acting as key risk transmitters. Portfolio simulations confirm the potential of clean energy assets and insurance ETFs in improving risk-adjusted returns. These findings underscore the importance of monitoring cross-sector interdependencies as the global energy transition reshapes the financial landscape.
- 28 April 2026
- Published by Elsevier in SSRN Electronic Journal
Open AccessBanking, Insurance Sector and Energy Transition: Volatility Spillovers Between U.S. and European Markets
This paper investigates the volatility spillovers between financial sector Exchange Trade Funds (ETFs) and energy markets, including fossil-based and clean energy assets, across U.S. and European markets. Using a TVP-VAR model, we quantify dynamic connectedness and assess systemic risk transmission, especially during periods of heightened uncertainty. Results show strong within-group connectedness in both financial and oil sectors, while clean energy emerges as a growing transmitter of shocks, particularly in the U.S. market. The analysis also highlights the distinct roles of financial institutions in volatility propagation, with European banks and insurers acting as key risk transmitters. Portfolio simulations confirm the potential of clean energy assets and insurance ETFs in improving risk-adjusted returns. These findings underscore the importance of monitoring cross-sector interdependencies as the global energy transition reshapes the financial landscape.
- Open Access
- 3 March 2026
- Published by Elsevier in Research in International Business and Finance
- Vol. 86
Energy and foreign exchanges market: Mapping risk and return connectedness in developed and emerging economies
- 3 March 2026
- Published by Elsevier in Research in International Business and Finance
- Vol. 86
Open AccessEnergy and foreign exchanges market: Mapping risk and return connectedness in developed and emerging economies
- Open Access
- 1 January 2026
- Published by University of Bialystok in Optimum. Economic Studies
- No. 3(125),p. 177-203
Prognozowanie ceny akcji spółki First Solar z wykorzystaniem uczenia maszynowego na podstawie wskaźników finansowych i makroekonomicznych
Purpose | The aim of this study is to assess the suitability of financial and macroeconomic indicators as predictors of First Solar’s short-term stock price in machine learning models. Two research questions were formulated: (1) which indicators demonstrate the strongest correlation with the stock price? and (2) which of the tested algorithms provides the highest forecast accuracy? Research method | The study analysed the relationship between First Solar’s stock price and selected financial and macroeconomic indicators from the first quarter of 2011 to the second quarter of 2024. Explanatory variables included financial indicators (profitability, market value, debt, and liquidity) and macroeconomic variables (CPI, GDP, and unemployment rate). Various machine learning algorithms were tested, including linear regression, Lasso regression, Ridge regression, decision trees, random forest, and ensemble models. Results | The study found that machine learning models are an effective tool for predicting stock prices. The ensemble model achieved the best results in forecasting First Solar’s stock price. Among the analysed metrics, profitability and market value had the greatest correlation with stock prices. The study confirmed that financial and macroeconomic indicators can be effective predictors for short-term stock price forecasting. Originality / value / implications / recommendations | This study contributes to the scientific literature by providing new evidence from the US stock market. The results have practical applications for market participants, such as traders and hedge funds, by providing tools for short-term stock price forecasting.
- 1 January 2026
- Published by University of Bialystok in Optimum. Economic Studies
- No. 3(125),p. 177-203
Open AccessPrognozowanie ceny akcji spółki First Solar z wykorzystaniem uczenia maszynowego na podstawie wskaźników finansowych i makroekonomicznych
Purpose | The aim of this study is to assess the suitability of financial and macroeconomic indicators as predictors of First Solar’s short-term stock price in machine learning models. Two research questions were formulated: (1) which indicators demonstrate the strongest correlation with the stock price? and (2) which of the tested algorithms provides the highest forecast accuracy? Research method | The study analysed the relationship between First Solar’s stock price and selected financial and macroeconomic indicators from the first quarter of 2011 to the second quarter of 2024. Explanatory variables included financial indicators (profitability, market value, debt, and liquidity) and macroeconomic variables (CPI, GDP, and unemployment rate). Various machine learning algorithms were tested, including linear regression, Lasso regression, Ridge regression, decision trees, random forest, and ensemble models. Results | The study found that machine learning models are an effective tool for predicting stock prices. The ensemble model achieved the best results in forecasting First Solar’s stock price. Among the analysed metrics, profitability and market value had the greatest correlation with stock prices. The study confirmed that financial and macroeconomic indicators can be effective predictors for short-term stock price forecasting. Originality / value / implications / recommendations | This study contributes to the scientific literature by providing new evidence from the US stock market. The results have practical applications for market participants, such as traders and hedge funds, by providing tools for short-term stock price forecasting.
- Open Access
- 24 November 2025
- Published by Springer Nature in Computational Economics
- Vol. 68 (5) , 1-25
Volatility Spillovers and Hedging Effectiveness of Green Bonds: A Cross-Regional Analysis
This research investigates the correlation transfer between green and conventional bonds across different geographical regions, highlighting the diverse approaches to climate policy in Europe, the US, and China. By employing a Time-Varying Parameter Vector Autoregressive (TVP-VAR) model, we analyze the phenomenon of risk contagion and cross-national interconnections in the bond market. Given the period of significant economic turbulence, our study spans from May 1, 2019, to December 31, 2023. Additionally, we apply multivariate portfolio techniques to evaluate the effectiveness of hedging strategies, and our findings show that green bonds have a greater tendency to transmit volatility than conventional bonds, underscoring the importance of adopting a cross-regional perspective to understand these dynamics. For investors, strategic allocation of green bonds, such as in a Minimum Variance Portfolio, may reduce portfolio volatility and serve as a hedge against market risk. Policymakers should address the vulnerabilities of green finance instruments and the interconnectedness of the global market when designing regulatory frameworks and risk mitigation strategies.
- 24 November 2025
- Published by Springer Nature in Computational Economics
- Vol. 68 (5) , 1-25
Open AccessVolatility Spillovers and Hedging Effectiveness of Green Bonds: A Cross-Regional Analysis
This research investigates the correlation transfer between green and conventional bonds across different geographical regions, highlighting the diverse approaches to climate policy in Europe, the US, and China. By employing a Time-Varying Parameter Vector Autoregressive (TVP-VAR) model, we analyze the phenomenon of risk contagion and cross-national interconnections in the bond market. Given the period of significant economic turbulence, our study spans from May 1, 2019, to December 31, 2023. Additionally, we apply multivariate portfolio techniques to evaluate the effectiveness of hedging strategies, and our findings show that green bonds have a greater tendency to transmit volatility than conventional bonds, underscoring the importance of adopting a cross-regional perspective to understand these dynamics. For investors, strategic allocation of green bonds, such as in a Minimum Variance Portfolio, may reduce portfolio volatility and serve as a hedge against market risk. Policymakers should address the vulnerabilities of green finance instruments and the interconnectedness of the global market when designing regulatory frameworks and risk mitigation strategies.
- 1 January 2025
- Published by University of Warsaw in Journal of Banking and Financial Economics
- Vol. 2025 (2(24)) , 113-134
Connectedness of Uncertainty, Volatility, and Stock Market Performance
This study examines the dynamic connectedness and spillover effects among various financial and economic indicators, including uncertainty indices, market volatility, and stock market indices, from 3 June 2008, to 30 December 2024. This interconnectedness implies that shocks originating in one market or asset class can rapidly transmit to others, underscoring the potential for systemic risk and financial contagion. The US emerges as a significant net transmitter of influence within the global financial system, with the VIX playing a crucial role in influencing global financial conditions. Major European equity markets also transmit influence, while the Geopolitical Risk Index (GRI) and the Economic Policy Uncertainty Index (EPUI) are primarily influenced by the broader financial system. Total connectedness varies over time, spiking during periods of economic distress, and volatility indices exhibit positive net connectedness during periods of financial stress, indicating their role as significant sources of volatility transmission.
- 1 January 2025
- Published by University of Warsaw in Journal of Banking and Financial Economics
- Vol. 2025 (2(24)) , 113-134
Connectedness of Uncertainty, Volatility, and Stock Market Performance
This study examines the dynamic connectedness and spillover effects among various financial and economic indicators, including uncertainty indices, market volatility, and stock market indices, from 3 June 2008, to 30 December 2024. This interconnectedness implies that shocks originating in one market or asset class can rapidly transmit to others, underscoring the potential for systemic risk and financial contagion. The US emerges as a significant net transmitter of influence within the global financial system, with the VIX playing a crucial role in influencing global financial conditions. Major European equity markets also transmit influence, while the Geopolitical Risk Index (GRI) and the Economic Policy Uncertainty Index (EPUI) are primarily influenced by the broader financial system. Total connectedness varies over time, spiking during periods of economic distress, and volatility indices exhibit positive net connectedness during periods of financial stress, indicating their role as significant sources of volatility transmission.
- Open Access
- 16 October 2024
- Published by Elsevier in Journal of Commodity Markets
- Vol. 36
Importance of geopolitical risk in volatility structure: New evidence from biofuels, crude oil, and grains commodity markets
- 16 October 2024
- Published by Elsevier in Journal of Commodity Markets
- Vol. 36
Open AccessImportance of geopolitical risk in volatility structure: New evidence from biofuels, crude oil, and grains commodity markets
- Top Cited PapersOpen Access
- 4 April 2023
- Published by Elsevier in Journal of International Financial Markets, Institutions and Money
- Vol. 85
How does the Russian-Ukrainian war change connectedness and hedging opportunities? Comparison between dirty and clean energy markets versus global stock indices
- 4 April 2023
- Published by Elsevier in Journal of International Financial Markets, Institutions and Money
- Vol. 85
Top Cited PapersOpen AccessHow does the Russian-Ukrainian war change connectedness and hedging opportunities? Comparison between dirty and clean energy markets versus global stock indices
- 1 January 2023
- Published by University of Warsaw
Dynamika powiązań i mechanizmy transmisji zmienności między polskim rynkiem akcji a wybranymi rynkami akcji na świecie
The monograph is a summary of the author's several years of research on the dynamics of linkages and volatility transmission mechanisms between the Polish stock market and selected stock markets around the world. It constitutes a valuable scientific position and a teaching aid for lecturers and students of economic faculties (such as banking and management, economics, finance and accounting, computer science and econometrics, management). It may also be of interest to financial market practitioners or individual investors who wish to better understand the interconnectedness of markets.
- 1 January 2023
- Published by University of Warsaw
Dynamika powiązań i mechanizmy transmisji zmienności między polskim rynkiem akcji a wybranymi rynkami akcji na świecie
The monograph is a summary of the author's several years of research on the dynamics of linkages and volatility transmission mechanisms between the Polish stock market and selected stock markets around the world. It constitutes a valuable scientific position and a teaching aid for lecturers and students of economic faculties (such as banking and management, economics, finance and accounting, computer science and econometrics, management). It may also be of interest to financial market practitioners or individual investors who wish to better understand the interconnectedness of markets.