Krystian Jaworski
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Krystian Jaworski has a total of 6 co-authors. Here are the top co-authors by number of shared publications:
- Jakub Borowski
Department of Economics II, SGH Warsaw School of Economics, Warsaw, Poland
5 shared publications - Jakub Olipra
SGH Warsaw School of Economics, Collegium of Management and Finance , Department of Applied Economics , ul. Madalińskiego 6/8 , Warsaw , Poland
2 shared publications - Jarko Fidrmuc
Zeppelin University Friedrichshafen, Friedrichshafen, Germany
2 shared publications Jakub Borowski
Collegium of World Economy, SGH Warsaw School of Economics , Warsaw , Poland
1 shared publicationM. Jahangir Alam
1 shared publication- Nemanja Popović
University of Montenegro
1 shared publication
Latest Publications
- Open Access
- 3 September 2026
- Published by Elsevier in SSRN Electronic Journal
Synthetic Consensus: Evaluating Large Language Model Forecasts with Repeated Draws
Asking a large language model the same question twice gives different answers, so a result based on one response mixes the model's underlying forecast with random variation between calls. We query the model 500 times for each quarterly inflation forecast in three economies, using target quarters that fall after the model's knowledge cutoff, and average the responses. Averaging lowers forecast error by 7 to 35 percent, and most of the gain arrives within ten repetitions. An exact decomposition separates the error of the average forecast from the removable variation between calls; that share is typically 16 percent of single-response error but reaches 76 percent, and it governs how much averaging helps. A more capable model varies less between calls and gains correspondingly less. Repeated querying should form part of any evaluation in which language model outputs are treated as data, and the number of repetitions should be reported.
- 3 September 2026
- Published by Elsevier in SSRN Electronic Journal
Open AccessSynthetic Consensus: Evaluating Large Language Model Forecasts with Repeated Draws
Asking a large language model the same question twice gives different answers, so a result based on one response mixes the model's underlying forecast with random variation between calls. We query the model 500 times for each quarterly inflation forecast in three economies, using target quarters that fall after the model's knowledge cutoff, and average the responses. Averaging lowers forecast error by 7 to 35 percent, and most of the gain arrives within ten repetitions. An exact decomposition separates the error of the average forecast from the removable variation between calls; that share is typically 16 percent of single-response error but reaches 76 percent, and it governs how much averaging helps. A more capable model varies less between calls and gains correspondingly less. Repeated querying should form part of any evaluation in which language model outputs are treated as data, and the number of repetitions should be reported.
- 8 April 2026
- Published by Wiley in Journal of Forecasting
- Vol. 45 (5) , 2525-2548
AI‐Driven Inflation Forecasting in the Aftermath of COVID‐19
This paper develops a hybrid approach to inflation forecasting that combines large language models (LLMs) with traditional econometric methods. We construct a novel indicator of inflation expectations by applying GPT‐4o to classify Twitter discussions among economists and financial commentators in the United States, the Eurozone, and Poland. Embedded in an ARIMAX framework, this indicator delivers systematic predictive content and consistently anticipates inflation dynamics up to 1 year ahead. Out‐of‐sample tests show that AI‐augmented forecasts achieve statistically significant accuracy gains over both professional consensus forecasts and benchmark ARIMA models, particularly at medium and long horizons. Robustness checks—including alternative smoothing and lag structures, model variants, prompt formulations, and metadata weighting—confirm the stability of results across regions and specifications. By integrating high‐frequency, sentiment‐based signals with econometric forecasting, the study demonstrates how generative AI can complement conventional tools, offering central banks, policymakers, and financial market participants an early‐warning mechanism during periods of uncertainty and structural change. While the approach is constrained by its reliance on Twitter data and the interpretability limits of LLMs, it highlights the transformative potential of hybrid AI‐econometric frameworks for real‐time macroeconomic analysis.
- 8 April 2026
- Published by Wiley in Journal of Forecasting
- Vol. 45 (5) , 2525-2548
AI‐Driven Inflation Forecasting in the Aftermath of COVID‐19
This paper develops a hybrid approach to inflation forecasting that combines large language models (LLMs) with traditional econometric methods. We construct a novel indicator of inflation expectations by applying GPT‐4o to classify Twitter discussions among economists and financial commentators in the United States, the Eurozone, and Poland. Embedded in an ARIMAX framework, this indicator delivers systematic predictive content and consistently anticipates inflation dynamics up to 1 year ahead. Out‐of‐sample tests show that AI‐augmented forecasts achieve statistically significant accuracy gains over both professional consensus forecasts and benchmark ARIMA models, particularly at medium and long horizons. Robustness checks—including alternative smoothing and lag structures, model variants, prompt formulations, and metadata weighting—confirm the stability of results across regions and specifications. By integrating high‐frequency, sentiment‐based signals with econometric forecasting, the study demonstrates how generative AI can complement conventional tools, offering central banks, policymakers, and financial market participants an early‐warning mechanism during periods of uncertainty and structural change. While the approach is constrained by its reliance on Twitter data and the interpretability limits of LLMs, it highlights the transformative potential of hybrid AI‐econometric frameworks for real‐time macroeconomic analysis.
- Open Access
- 16 December 2025
- Published by Springer Nature in Computational Economics
- Vol. 68 (5) , 1-31
Measuring Inflation Expectations Using Artificial Intelligence
- 16 December 2025
- Published by Springer Nature in Computational Economics
- Vol. 68 (5) , 1-31
Open AccessMeasuring Inflation Expectations Using Artificial Intelligence
- Open Access
- 20 October 2025
- Published by Warsaw University of Life Sciences - SGGW Press in Metody Ilościowe w Badaniach Ekonomicznych
- Vol. 26 (2) , 60-69
Forecasting in Turbulent Times: How Artificial Intelligence and Machine Learning Are Reshaping Macroeconomic Prediction
This paper examines how artificial intelligence and machine learning have reshaped macroeconomic forecasting in the volatile post-COVID era. Highlighting the use of ensemble methods, neural networks, and large language models, it illustrates their advantages in capturing nonlinear dynamics and processing complex data. Drawing on central bank case studies, the paper shows that AI enhances predictive power, though interpretability and robustness remain challenges. AI is best seen as a complement to, not a replacement for, traditional economic models and human judgment.
- 20 October 2025
- Published by Warsaw University of Life Sciences - SGGW Press in Metody Ilościowe w Badaniach Ekonomicznych
- Vol. 26 (2) , 60-69
Open AccessForecasting in Turbulent Times: How Artificial Intelligence and Machine Learning Are Reshaping Macroeconomic Prediction
This paper examines how artificial intelligence and machine learning have reshaped macroeconomic forecasting in the volatile post-COVID era. Highlighting the use of ensemble methods, neural networks, and large language models, it illustrates their advantages in capturing nonlinear dynamics and processing complex data. Drawing on central bank case studies, the paper shows that AI enhances predictive power, though interpretability and robustness remain challenges. AI is best seen as a complement to, not a replacement for, traditional economic models and human judgment.
- Open Access
- 25 September 2025
- Published by Elsevier in Economics Letters
- Vol. 257
Artificial intelligence and inflation in the EU
- 25 September 2025
- Published by Elsevier in Economics Letters
- Vol. 257
Open AccessArtificial intelligence and inflation in the EU
- 5 August 2025
- Published by Elsevier in Structural Change and Economic Dynamics
- Vol. 75, 213-227
The impact of excess household savings on inflation dynamics in the EU during the COVID-19 era
- 5 August 2025
- Published by Elsevier in Structural Change and Economic Dynamics
- Vol. 75, 213-227
The impact of excess household savings on inflation dynamics in the EU during the COVID-19 era
- Open Access
- 4 March 2025
- Published by Springer Nature in Empirica
- Vol. 52 (3) , 413-433
Convergence, inequality and inflation synchronization: evidence from the Eurozone
This paper studies the impact of real convergence in the EU countries on inflation synchronization between these countries and the Eurozone. Inflation co-movement between the Eurozone and the EU countries serves as an important measure of the adequacy of the single monetary policy for both current and future members of the common currency area. We report three major results. First, countries with higher relative GDP per capita in the EU countries report stronger inflation co-movement. Second, the relationship between real convergence and the inflation synchronization is non-linear. Third, lower income inequality is associated with greater inflation co-movement. Our findings suggest that real convergence in the EU is associated with stronger inflation synchronization between the EU countries and Eurozone and more effective common monetary policy in the long run. These results provide support for the “coronation theory” which underscores that monetary integration should follow, rather than precede, the process of real convergence. We show that for the catching-up countries with relatively high GDP per capita the value of waiting for the income gap to narrow is limited as additional convergence implies only moderate increase in inflation co-movement.
- 4 March 2025
- Published by Springer Nature in Empirica
- Vol. 52 (3) , 413-433
Open AccessConvergence, inequality and inflation synchronization: evidence from the Eurozone
This paper studies the impact of real convergence in the EU countries on inflation synchronization between these countries and the Eurozone. Inflation co-movement between the Eurozone and the EU countries serves as an important measure of the adequacy of the single monetary policy for both current and future members of the common currency area. We report three major results. First, countries with higher relative GDP per capita in the EU countries report stronger inflation co-movement. Second, the relationship between real convergence and the inflation synchronization is non-linear. Third, lower income inequality is associated with greater inflation co-movement. Our findings suggest that real convergence in the EU is associated with stronger inflation synchronization between the EU countries and Eurozone and more effective common monetary policy in the long run. These results provide support for the “coronation theory” which underscores that monetary integration should follow, rather than precede, the process of real convergence. We show that for the catching-up countries with relatively high GDP per capita the value of waiting for the income gap to narrow is limited as additional convergence implies only moderate increase in inflation co-movement.
- Open Access
- 20 February 2025
- Published by Elsevier in Food Policy
- Vol. 131
Cutting VAT rate on food products in a high-inflation environment. Does it work out?
- 20 February 2025
- Published by Elsevier in Food Policy
- Vol. 131
Open AccessCutting VAT rate on food products in a high-inflation environment. Does it work out?
- Open Access
- 1 March 2024
- Published by Walter de Gruyter GmbH in International Journal of Management and Economics
- Vol. 60 (1) , 33-45
The Russian invasion of Ukraine and the exchange rate of the Polish zloty: A fallacy of monetary autonomy?
We use the Twitter application programming interface to construct a novel indicator capturing the varying perceptions of geopolitical risk related to the war in Ukraine. We show that the Twitter variable is a statistically significant determinant of the EUR/PLN exchange rate following the invasion. We estimate that the war in Ukraine was responsible for an increase in the EUR/PLN exchange rate of about 5.0% in the first 2 weeks following its breakout and that the weakening of PLN contributed to an increase of headline inflation in Poland by about 0.33 percentage points. Our findings suggest that monetary policy autonomy can be substantially constrained due to conflicts in nearby countries. Therefore, they provide an argument for Poland in seeking Eurozone membership as a way to contain the impact of geopolitical factors on exchange rate volatility, which is likely to remain at play in the longer run.
- 1 March 2024
- Published by Walter de Gruyter GmbH in International Journal of Management and Economics
- Vol. 60 (1) , 33-45
Open AccessThe Russian invasion of Ukraine and the exchange rate of the Polish zloty: A fallacy of monetary autonomy?
We use the Twitter application programming interface to construct a novel indicator capturing the varying perceptions of geopolitical risk related to the war in Ukraine. We show that the Twitter variable is a statistically significant determinant of the EUR/PLN exchange rate following the invasion. We estimate that the war in Ukraine was responsible for an increase in the EUR/PLN exchange rate of about 5.0% in the first 2 weeks following its breakout and that the weakening of PLN contributed to an increase of headline inflation in Poland by about 0.33 percentage points. Our findings suggest that monetary policy autonomy can be substantially constrained due to conflicts in nearby countries. Therefore, they provide an argument for Poland in seeking Eurozone membership as a way to contain the impact of geopolitical factors on exchange rate volatility, which is likely to remain at play in the longer run.
- Open Access
- 9 January 2023
- Published by Springer Nature in Empirica
- Vol. 50 (1) , 237-253
Economic and behavioral determinants of forced household savings during the COVID-19 pandemic
This study seeks to identify the determinants of forced household savings in 16 European Union (EU) member states in 2020. We show that the higher the severity of the COVID-19 pandemic in the state, measured by the intensity of government restrictions or the number of COVID-19-related deaths, the higher the level of forced savings. Such savings also increased with gross domestic product per capita and the financial support provided for households and enterprises by the government. Additionally, savings cultures and personality traits that support compliance with pandemic-related restrictions and enhance coping with the hardship of the pandemic had a positive impact on forced savings. Our results show that while common pandemic shock may lead to discrepancies in forced savings in affected countries, their level depends largely on government response in the form of imposed restrictions as well as financial support for households and enterprises. Therefore, strong fiscal support during the pandemic can be likened to sowing the seeds for post-pandemic recovery, as savings accumulated during the pandemic shock may be used to finance the pent-up demand. This, in turn, suggests that fiscal responses during the pandemic may act as a significant driver of post-pandemic business cycle (de)synchronization and inflation differentials among EU member states and, more importantly, euro-area countries.
- 9 January 2023
- Published by Springer Nature in Empirica
- Vol. 50 (1) , 237-253
Open AccessEconomic and behavioral determinants of forced household savings during the COVID-19 pandemic
This study seeks to identify the determinants of forced household savings in 16 European Union (EU) member states in 2020. We show that the higher the severity of the COVID-19 pandemic in the state, measured by the intensity of government restrictions or the number of COVID-19-related deaths, the higher the level of forced savings. Such savings also increased with gross domestic product per capita and the financial support provided for households and enterprises by the government. Additionally, savings cultures and personality traits that support compliance with pandemic-related restrictions and enhance coping with the hardship of the pandemic had a positive impact on forced savings. Our results show that while common pandemic shock may lead to discrepancies in forced savings in affected countries, their level depends largely on government response in the form of imposed restrictions as well as financial support for households and enterprises. Therefore, strong fiscal support during the pandemic can be likened to sowing the seeds for post-pandemic recovery, as savings accumulated during the pandemic shock may be used to finance the pent-up demand. This, in turn, suggests that fiscal responses during the pandemic may act as a significant driver of post-pandemic business cycle (de)synchronization and inflation differentials among EU member states and, more importantly, euro-area countries.