|
on Computational Economics |
| By: | Leonardo Ciambezi (Université Côte D’Azur, CNRS, GREDEG and Institute of Economics and l’Embeds, Scuola Superiore Sant’Anna di Pisa); Mattia Guerini (Department of Economics and Management, University of Brescia, Fondazione Eni Enrico Mattei, Université Côte D’Azur, CNRS, GREDEG and Institute of Economics and l’Embeds, Scuola Superiore Sant’Anna di Pisa); Mauro Napoletano (Université Côte D’Azur, CNRS, GREDEG, OFCE - SciencesPo and Institute of Economics and l’Embeds, Scuola Superiore Sant’Anna di Pisa); Andrea Roventini (Institute of Economics and l’Embeds, Scuola Superiore Sant’Anna di Pisa and OFCE - SciencesPo) |
| Abstract: | We develop a macroeconomic agent-based model to study the role of demand and supply factors in determining inflation dynamics. Local interactions of heterogeneous firms and households in the labor and goods markets characterize the model. Asymmetric information implies that firm selection is imperfect and depends both on firms’ relative prices and on their size. We calibrate the model on EU data by using the method of simulated moments and show that it can generate realistic inflation dynamics and a non-linear Phillips curve in line with recent empirical evidence. We then find that the traditional demand-led explanation of inflation stemming from a tight labor market only holds when selection in the goods markets is mostly driven by relative prices in comparison to firm size. Finally, we study the response of inflation to shocks impacting consumption, labor productivity, or energy costs. The results indicate that only demand shocks lead to wage-led inflation surges. Productivity shocks are entirely passed through to prices without affecting the income distribution. Energy shocks, instead, induce sellers’ inflation after changes in both firms’ cost structure and profit margins. This is in line with the recent empirical evidence for the Euro Area. |
| Keywords: | Inflation, agent-based models, market structure, mark-up rates, sellers’ inflation |
| JEL: | E31 E32 C63 |
| Date: | 2025–03 |
| URL: | https://d.repec.org/n?u=RePEc:fem:femwpa:2025.10 |
| By: | Maddalena Honorati; Céline Ferré; Tomasz Gajderowicz |
| Keywords: | Social Protections and Labor-Labor Markets Information and Communication Technologies-ICT Applications |
| Date: | 2023–11 |
| URL: | https://d.repec.org/n?u=RePEc:wbk:wboper:40628 |
| By: | Iftikhar, Ilaria Giannoccaro & Anas |
| Abstract: | The ripple effect refers to disruption propagation across the supply network affecting its global performance. To cope with it, supply networks should be resilient. This study investigates the drivers of supply network resilience, viewed as adaptive capacity to disruptions, focusing on trust and investigating the moderating role of network topology on the relationship between trust and resilience. We first develop an NK agent-based model of the supply network to simulate resilient performance. Then, a simulation analysis is carried out, to assess the effect of trust on the resilience of supply networks displaying different complex topologies. Our results confirm that trust positively affects supply network resilience; however, across the different topologies, the beneficial effect of trust varies. In particular, we find that trust is beneficial at most for the following topologies: local, small-world, block-diagonal, and random. For centralised, diagonal, and hierarchical topologies improving trust increases resilience at a moderate level. We also find that, as the frequency of disruptions rises, the positive effect of trust on resilience decreases. Managerial implications of the main findings are finally discussed |
| Date: | 2023–06–26 |
| URL: | https://d.repec.org/n?u=RePEc:osf:osfxxx:2spt3_v1 |
| By: | Bokan, Nikola; Lenza, Michele; Araujo, Douglas; Comazzi, Fabio Alberto |
| Abstract: | Word embeddings are vectors of real numbers associated with words, designed to capture semantic and syntactic similarity between the words in a corpus of text. We estimate the word embeddings of the European Central Bank’s introductory statements at monetary policy press conferences by using a simple natural language processing model (Word2Vec), only based on the information and model parameters available as of each press conference. We show that a measure based on such embeddings contributes to improve core inflation forecasts multiple quarters ahead. Other common textual analysis techniques, such as dictionary-based metrics or sentiment metrics do not obtain the same results. The information contained in the embeddings remains valuable for out-of-sample forecasting even after controlling for the central bank inflation forecasts, which are an important input for the introductory statements. JEL Classification: E31, E37, E58 |
| Keywords: | central bank texts, embeddings, forecasting, inflation |
| Date: | 2025–04 |
| URL: | https://d.repec.org/n?u=RePEc:ecb:ecbwps:20253047 |
| By: | Heng-fu Zou |
| Abstract: | This study models the cyclical rise and fall of Chinese dynasties through the interplay of power and corruption using the Van der Pol equation. It captures the historical pattern where dynasties emerge with low corruption and rising power, reach a peak, and collapse as corruption undermines authority. The Van der Pol model, known for generating self-sustaining cycles, reflects how early dynastic reforms drive rapid power growth, while increasing corruption accelerates decline and collapse. The system then resets, mirroring the continuous rise and fall observed in Chinese history. Additionally, the study analyzes the evolution of the system's total energy, revealing how power and corruption interact to drive cycles of expansion, instability, and collapse. Simulations further show how cor ruption feedback intensity influences the length and stability of dynastic cycles, explaining why some dynasties, like the Han and Tang, endured longer than others, such as the Song and Ming. This research offers a quantitative framework to understand the repetitive nature of dynastic cycles driven by power and corruption. |
| Date: | 2025–03–13 |
| URL: | https://d.repec.org/n?u=RePEc:cuf:wpaper:747 |
| By: | Torój , Andrzej (SGH Warsaw School of Economics); Bęza-Bojanowska, Joanna (Ministry of Finance); Chmura, Rafał (Ministry of Finance); Kroschel, Dominika (Ministry of Finance); Szczypińska, Agnieszka (Ministry of Finance); Wiśnicki, Bartłomiej (SGH Warsaw School of Economics) |
| Abstract: | We investigate the properties of the Polish numerical stabilizing expenditure rule (SER) in the context of economic governance review in the EU. To this aim, we use the macroeconometric model NEMPF (Chmura et al., 2024) that offers nuanced, disaggregated mapping between the general government (GG) expenditure categories, the macroeconomic variables (including GDP), and GG revenue categories. This set of detailed links allows for heterogeneous fiscal multipliers by expenditure category, and hence scenario-specific calculation of our categories of interest: the GG revenue, expenditure and balance developments as ratios to GDP. The model-based, endogenous denominator properly accounts for tax base and hence revenue responses to expenditure-side measures. As the Polish SER represents a forward-looking perspective, we propose model solution procedures under model-consistent expectations of policymakers applicable when the perfect foresight assumption is not met. We find that SER generally ensures lower GG deficits (and hence GG debt paths) than policies targeting just-compliance with the Stability and Growth Pact (SGP) thresholds. That results in lower GG debt trajectories, as well as creates room for counter-cyclical responses. We also demonstrate a few specific numerical properties of the Polish SER, including how the correction mechanism encourages a more restrictive fiscal policy to build countercyclical buffers. |
| Keywords: | fiscal rules; macroeconomic simulation; public finance; European Union |
| JEL: | C54 E62 H30 |
| Date: | 2025–03–31 |
| URL: | https://d.repec.org/n?u=RePEc:ris:mfplwp:0044 |
| By: | Hwang, Minho; Chung, Dongil |
| Abstract: | Classical decision theory assumes a central valuation system in which the brain encodes subjective values-or utilities-for all available options in a given choice set, regardless of how different the options are (common currency). Garcia et al. (2023) pointed out that although there has been experimental evidence for differential properties between experience-based (experiential) and description-based (symbolic) choices, the alternative possibility suggesting the existence of separate valuation systems for each modality has not been directly assessed. The authors reported empirical results supporting the alternative hypothesis that participants recruit different valuation systems for each modality. Here, we reproduced the results of this original paper and performed robustness checks. Overall, we reproduced most of the statistical results and model-based results of the original study. We employed two additional methods to test the robustness of the computational modeling used in the original study: parameterrecovery using the scripts shared by the authors and parameter estimations using different model fitting methods (maximum log-likelihood estimation (MLE) and hierarchical Bayesian estimation). Our parameter-recovery method successfully recovered most of the original model parameters, estimated from choice between experiential and symbolic values (ES phase) and between two experiential values (EE phase). Through consecutive analyses, including alternative parameter estimation methods, we confirmed that the issue does not compromise the original study's conclusions, and that all results directly related to the main conclusion (i.e., indifference points) are reproducible." |
| Date: | 2025 |
| URL: | https://d.repec.org/n?u=RePEc:zbw:i4rdps:215 |