arxiv: 2605.10447 · v1 · submitted 2026-05-11 · 💻 cs.MA · cs.AI· econ.GN· q-fin.EC· q-fin.ST

Recognition: no theorem link

Statistical Model Checking of the Keynes+Schumpeter Model: A Transient Sensitivity Analysis of a Macroeconomic ABM

Stefano Blando , Giorgio Fagiolo , Mauro Napoletano , Tania Treibich , Andrea Vandin

Authors on Pith no claims yet

Pith reviewed 2026-05-12 04:40 UTC · model grok-4.3

classification 💻 cs.MA cs.AIecon.GNq-fin.ECq-fin.ST

keywords agent-based modelsstatistical model checkingKeynes-Schumpeter modelmacroeconomic ABMtransient sensitivity analysisunemploymentGDP growthsimulation stopping rules

0 comments

The pith

Statistical model checking delivers reproducible transient sensitivity analysis for the Keynes-Schumpeter macroeconomic agent-based model.

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The paper shows that statistical model checking can replace ad hoc Monte Carlo campaigns when studying complex macroeconomic agent-based models. It applies the method to the heuristic-switching Keynes+Schumpeter model through one-parameter sweeps on unemployment, GDP growth, and an auxiliary market-share probe over the post-warmup segment of a 600-step horizon. Reusable temporal queries together with observable-specific precision targets and confidence-based stopping rules automatically set the simulation effort required for each configuration. The resulting data reveal that macro-financial and structural sweeps generate the largest transient effects while several heuristic-rule sweeps stay weaker under the same precision policy. A reader cares because the approach makes uncertainty estimates and simulation cost explicit parts of the reported results and works on the original simulator code without rewriting.

Core claim

Statistical model checking supports quantitative transient sensitivity analysis of the Keynes+Schumpeter model by driving simulation runs with temporal queries, precision targets, and automatic stopping rules that adapt effort to each parameter setting and observable. Across the tested sweeps the method identifies clear differences: macro-financial and structural parameters produce the strongest short-term changes in unemployment and GDP growth, whereas heuristic-rule parameters yield comparatively muted transients under identical statistical guarantees.

What carries the argument

Statistical model checking driven by reusable temporal queries, observable-specific precision targets, and confidence-based stopping rules applied to the post-warmup phase of the heuristic-switching Keynes+Schumpeter agent-based simulator.

If this is right

Macro-financial and structural parameter sweeps produce stronger transient effects on unemployment and GDP growth than heuristic-rule sweeps under the same precision policy.
Uncertainty estimates and the simulation effort required become explicit and comparable across all tested configurations.
Analysis of substantively rich economic agent-based models can proceed without rewriting the simulator in a dedicated formal language.
Transient rather than steady-state focus yields new distinctions among parameter families that steady-state Monte Carlo campaigns can miss.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

The same query-and-precision framework could be reused on other macroeconomic agent-based models to produce comparable sensitivity rankings.
Policy-relevant short-term fluctuation analysis could be strengthened by systematically varying the post-warmup horizon length to test persistence of the reported contrasts.
Market-share micro-probes could be expanded to additional firm-level statistics to check whether the macro contrasts are accompanied by consistent micro-level shifts.

Load-bearing premise

The selected temporal queries, precision targets, and 600-step post-warmup window are sufficient to capture the relevant transient dynamics of the model.

What would settle it

A finding that extending the simulation horizon beyond 600 steps or altering the precision targets reverses the ranking of which parameter families produce the strongest transients would show that the chosen analysis window misses important behavior.

Figures

Figures reproduced from arXiv: 2605.10447 by Andrea Vandin, Giorgio Fagiolo, Mauro Napoletano, Stefano Blando, Tania Treibich.

**Figure 1.** Figure 1: Cross-experiment scorecard of the transient sensitivity campaign. Each row summarizes one sweep by trade-off counts, counterfactual separation, best tail signal, and mean sample complexity across the three observables. while market share responds selectively, most notably under ‘E12’ and ‘E10’. Full values appear in the appendix tables. 6.1 Switching Parameters (E1–E3) The direct switching sweeps are the w… view at source ↗

**Figure 2.** Figure 2: Sensitivity summary for ‘E4’ (τ , tax rate). Unemployment bands fan out early and monotonically, while GDP growth and market share remain more weakly separated. 6.3 Heuristic Rule Coefficients (E5–E8) The rule-internal coefficients ‘E5’–‘E8’ often separate unemployment trajectories — ‘E5’ reaches 12 of 15 final pairs and ‘E6’ reaches 10 — but remain weak or intermittent for GDP growth and market share. The… view at source ↗

**Figure 3.** Figure 3: Sensitivity summary for ‘E9’ (τb, credit parameter). Unemployment bands diverge sharply, GDP growth separates more modestly, and market-share effects remain limited. indicating that tighter credit requirements simultaneously raise unemployment and reduce GDP growth [PITH_FULL_IMAGE:figures/full_fig_p013_3.png] view at source ↗

**Figure 4.** Figure 4: Sensitivity summary for ‘E10’ (µdeb, initial bank markup on loan rates). This is the clearest directional win-win case: unemployment improves strongly, GDP growth shifts in a favorable direction, and market share also moves across the sweep. evidence that GDP growth is generally easy to separate under this campaign design. Inventory expectations ι (E12). The inventory-expectation sweep is the strongest mar… view at source ↗

**Figure 5.** Figure 5: Sensitivity summary for ‘E12’ (ι, inventory expectations). This is the strongest market-share case and the costliest sweep in the campaign. Pareto heterogeneity a (E11). The Pareto-shape sweep is an outlier in the opposite direction. Despite covering a broad range of firm-size heterogeneity, it achieves only moderate unemployment separation (6 of 15 final pairs) and very limited response in the other two … view at source ↗

**Figure 6.** Figure 6: Sensitivity summary for ‘E1’ (β, intensity of choice). Trajectory bands overlap almost entirely for all three observables, with at most one significant unemployment pair at the final step [PITH_FULL_IMAGE:figures/full_fig_p024_6.png] view at source ↗

**Figure 7.** Figure 7: Sensitivity summary for ‘E2’ (δs, switching inertia). Similar to ‘E1’: no unemployment pair achieves significance at the final step, and GDP growth and market share remain flat. Heuristic Coefficients The rule-internal coefficients ‘E5’–‘E8’ produce mixed results. Unemployment is often separated, but GDP growth and market share are rarely moved by these parameters. Financial and Structural Parameters The … view at source ↗

**Figure 8.** Figure 8: Sensitivity summary for ‘E3’ (η, switching memory). Moderately stronger than ‘E1’ and ‘E2’ for unemployment (3 of 15 final pairs), but GDP growth and market share remain largely indistinguishable [PITH_FULL_IMAGE:figures/full_fig_p026_8.png] view at source ↗

**Figure 9.** Figure 9: Sensitivity summary for ‘E5’ (ωada, adaptive-expectations coefficient). Unemployment separation is strong (12 of 15 final pairs), but GDP growth and market share show limited response [PITH_FULL_IMAGE:figures/full_fig_p027_9.png] view at source ↗

**Figure 10.** Figure 10: Sensitivity summary for ‘E6’ (ωwtr, weak trend-following coefficient). Unemployment separation is notable (10 of 15 final pairs) but market share responds only weakly (4 of 15) [PITH_FULL_IMAGE:figures/full_fig_p028_10.png] view at source ↗

**Figure 11.** Figure 11: Sensitivity summary for ‘E7’ (ωstr, strong trend-following coefficient). Unemployment reaches 10 of 15 final pairs; market share shows moderate response (5 of 15), making this the third most market-share-reactive sweep after ‘E12’ and ‘E10’ [PITH_FULL_IMAGE:figures/full_fig_p029_11.png] view at source ↗

**Figure 12.** Figure 12: Sensitivity summary for ‘E8’ (ωaa, anchor-and-adjustment weight). Unemployment separation is moderate (5 of 15 final pairs). This is also the only heuristic sweep with two “win-win” points against the baseline [PITH_FULL_IMAGE:figures/full_fig_p030_12.png] view at source ↗

**Figure 13.** Figure 13: Sensitivity summary for ‘E11’ (a, Pareto shape). Moderate unemployment separation (6 of 15 final pairs) but very limited GDP-growth and market-share response, placing this sweep among the weakest in the financial and structural block [PITH_FULL_IMAGE:figures/full_fig_p031_13.png] view at source ↗

read the original abstract

Agent-based models (ABMs) are increasingly used in macroeconomics, but their analysis still often relies on ad hoc Monte Carlo campaigns with heterogeneous statistical effort across parameter settings. We show how statistical model checking (SMC), implemented through MultiVeStA, can provide a principled analysis layer for a realistic macroeconomic ABM without rewriting the simulator in a dedicated formalism. Our case study is the heuristic-switching Keynes+Schumpeter(K+S) model, analysed hrough a transient sensitivity campaign over one-parameter sweeps, two macro observables (unemployment and GDP growth), and one auxiliary micro-level probe (market share) on the post-warmup phase of a 600-step horizon. The analysis is driven by reusable temporal queries, observable-specific precision targets, and confidence-based stopping rules that automatically determine the simulation effort required by each configuration. Results show a clear contrast across parameter families: macro-financial and structural sweeps produce the strongest transient effects, whereas several heuristic-rule sweeps remain much weaker under the same precision policy. More broadly, the paper shows that SMC can support reproducible and informative quantitative analysis of substantively rich economic ABMs, while making uncertainty estimates and simulation cost explicit parts of the reported results.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Desk Editor's Note private letter to a colleague

SMC via MultiVeStA gives ABM sensitivity runs automatic stopping and explicit uncertainty on the K+S model, but the fixed 600-step post-warmup window still needs validation against longer transients.

read the letter

The paper shows how to attach statistical model checking through MultiVeStA to an existing macroeconomic ABM without rewriting it in a special language. They run transient sensitivity sweeps on the heuristic-switching K+S model, tracking unemployment, GDP growth, and market share over one-parameter families after a 600-step warmup. Reusable temporal queries plus precision targets and confidence-based stopping rules set the number of runs per configuration on the fly. The main payoff is a clear split: macro-financial and structural parameters produce stronger transient effects than most heuristic-rule changes under the same precision policy. That contrast is new for this model and makes simulation cost and statistical uncertainty part of the reported results rather than hidden in ad-hoc Monte Carlo choices. The approach is straightforward to reuse on other ABMs that already have a working simulator. The 600-step post-warmup horizon is the weakest part. The abstract gives no convergence diagnostics, no comparison to longer runs, and no check that dominant transients have settled for every parameter regime. If some sweeps still show non-monotonic behavior or slow adjustment after step 600, the reported differences could shrink or shift. Exact query formulations and any data-exclusion rules are also not spelled out, so it is hard to judge whether the MultiVeStA layer fully preserves the original model dynamics. This work is aimed at macro ABM users who already run Monte Carlo experiments and want a more controlled, reproducible layer on top. It deserves serious referee time because it addresses a real, recurring problem in the subfield with a concrete, reusable example. Ask the authors for horizon validation and query details in revision, but the core idea is worth reviewing.

Referee Report

1 major / 1 minor

Summary. The manuscript presents an application of statistical model checking (SMC) using the MultiVeStA tool to analyze the transient dynamics of the Keynes+Schumpeter (K+S) agent-based macroeconomic model. Through one-parameter sweeps on macro-financial, structural, and heuristic parameters, it examines sensitivity of unemployment, GDP growth, and market share over a fixed 600-step post-warmup horizon using temporal queries with precision targets and confidence-based stopping rules. The key finding is a contrast in the strength of transient effects across parameter families, with macro-financial and structural sweeps showing stronger effects than heuristic-rule sweeps. The paper argues that this approach enables reproducible quantitative analysis of rich ABMs while explicitly accounting for uncertainty and computational cost.

Significance. If the results hold, this work is significant in demonstrating how SMC can be integrated with existing ABM simulators without requiring a dedicated formal language, addressing the common issue of ad hoc Monte Carlo analyses in macroeconomic ABMs. It provides a method that makes simulation effort and statistical confidence explicit, which could improve reproducibility in the field. The contrast in results across parameter types offers substantive insight into the model's behavior, though its robustness depends on the adequacy of the chosen analysis horizon.

major comments (1)

Abstract: the analysis is described as covering the 'post-warmup phase of a 600-step horizon' with no mention of convergence diagnostics, comparison to longer runs, or analysis of the model's characteristic timescales. This choice is load-bearing for the central claim, as truncation bias in regimes where transients extend beyond 600 steps could artifactually produce the reported contrasts between macro-financial/structural and heuristic-rule parameter families.

minor comments (1)

The abstract refers to 'reusable temporal queries' and 'observable-specific precision targets' without specifying their exact formulations or how MultiVeStA integration preserves the original model dynamics; adding these details would strengthen reproducibility claims.

Simulated Author's Rebuttal

1 responses · 0 unresolved

We thank the referee for the constructive and detailed review. We address the single major comment below and outline revisions that will strengthen the manuscript while preserving the integrity of the reported analysis.

read point-by-point responses

Referee: [—] Abstract: the analysis is described as covering the 'post-warmup phase of a 600-step horizon' with no mention of convergence diagnostics, comparison to longer runs, or analysis of the model's characteristic timescales. This choice is load-bearing for the central claim, as truncation bias in regimes where transients extend beyond 600 steps could artifactually produce the reported contrasts between macro-financial/structural and heuristic-rule parameter families.

Authors: We acknowledge that the abstract and main text do not explicitly discuss convergence diagnostics or comparisons to longer horizons. The 600-step post-warmup window was selected to isolate transient dynamics, consistent with the simulation lengths used in prior K+S studies for capturing short-to-medium-term macroeconomic adjustments. Although formal convergence checks and extended runs are absent from the current manuscript, the SMC procedure applies identical precision targets and stopping rules across all parameter sweeps, ensuring that the reported contrasts reflect statistically controlled differences within the studied interval rather than uncontrolled Monte Carlo variation. The relative weakness of heuristic-rule effects is therefore observed under uniform conditions. We will revise the abstract to note the horizon's grounding in the model's established timescales and add a concise methods subsection (or discussion paragraph) that (i) references the characteristic adjustment periods from the K+S literature, (ii) states the absence of explicit truncation diagnostics as a limitation, and (iii) clarifies that future extensions could test longer horizons. These textual changes will make the analysis more transparent without requiring re-execution of the existing SMC campaigns. revision: yes

Circularity Check

0 steps flagged

No significant circularity; empirical application of external tool

full rationale

The paper applies statistical model checking via the external MultiVeStA tool to an existing macroeconomic ABM for transient sensitivity analysis over parameter sweeps. All results derive from simulation outputs and reusable temporal queries with confidence-based stopping rules rather than any derivations, equations, or fitted parameters that reduce to inputs by construction. No self-definitional steps, predictions forced by fits, or load-bearing self-citations appear in the described chain; the 600-step horizon is a fixed methodological parameter, not a circular element.

Axiom & Free-Parameter Ledger

0 free parameters · 1 axioms · 0 invented entities

The analysis rests on the correctness of the pre-existing K+S model implementation and the MultiVeStA tool; no new free parameters, axioms, or invented entities are introduced beyond standard assumptions of statistical model checking and agent-based simulation.

axioms (1)

domain assumption The K+S model implementation faithfully represents the intended macroeconomic dynamics.
The paper applies SMC to the existing model without re-deriving or verifying its internal rules.

pith-pipeline@v0.9.0 · 5543 in / 1201 out tokens · 31671 ms · 2026-05-12T04:40:36.467354+00:00 · methodology

discussion (0)

Reference graph

Works this paper leans on

48 extracted references · 48 canonical work pages

[1]

ACM Transactions on Modeling and Computer Simulation28(1), 1–39 (2018)

Agha, G., Palmskog, K.: A survey of statistical model checking. ACM Transactions on Modeling and Computer Simulation28(1), 1–39 (2018). https://doi.org/10. 1145/3158668

work page 2018
[2]

Econo- metrica60(2), 323–351 (1992)

Aghion, P., Howitt, P.: A model of growth through creative destruction. Econo- metrica60(2), 323–351 (1992). https://doi.org/10.2307/2951599

work page doi:10.2307/2951599 1992
[3]

American Economic Journal: Microeconomics4(4), 35–64 (2012)

Anufriev, M., Hommes, C.: Evolutionary selection of individual expectations and aggregate outcomes in asset pricing experiments. American Economic Journal: Microeconomics4(4), 35–64 (2012). https://doi.org/10.1257/mic.4.4.35, https:// www.aeaweb.org/articles?id=10.1257/mic.4.4.35

work page doi:10.1257/mic.4.4.35 2012
[4]

Journal of Evolutionary Eco- nomics23(3), 663–688 (2013)

Anufriev, M., Hommes, C.H., Philipse, R.H.S.: Evolutionary selection of expec- tations in positive and negative feedback markets. Journal of Evolutionary Eco- nomics23(3), 663–688 (2013). https://doi.org/10.1007/s00191-011-0242-4, https: //doi.org/10.1007/s00191-011-0242-4

work page doi:10.1007/s00191-011-0242-4 2013
[5]

The Review of Eco- nomic Studies29(3), 155–173 (1962)

Arrow, K.J.: The economic implications of learning by doing. The Review of Eco- nomic Studies29(3), 155–173 (1962). https://doi.org/10.2307/2295952

work page doi:10.2307/2295952 1962
[6]

MIT Press (2008)

Baier, C., Katoen, J.: Principles of model checking. MIT Press (2008). https://doi. org/10.1093/comjnl/bxp025

work page doi:10.1093/comjnl/bxp025 2008
[7]

In: Proceedings of MARS@ETAPS 2026 (2026)

Blando, S., Fagiolo, G., Giachini, D., Vandin, A., Ivanaj, E.: Statistical model checking of the island model: An established economic agent-based model of en- dogenous growth. In: Proceedings of MARS@ETAPS 2026 (2026). https://doi.org/ 10.4204/EPTCS.443.2

work page doi:10.4204/eptcs.443.2 2026
[8]

Econo- metrica65(5), 1059–1095 (1997)

Brock, W.A., Hommes, C.H.: A rational route to randomness. Econo- metrica65(5), 1059–1095 (1997). https://doi.org/10.2307/2171879, https://www.econometricsociety.org/publications/econometrica/1997/09/01/ rational-route-randomness

work page doi:10.2307/2171879 1997
[9]

Journal of Economic Dynamics and Control69, 375–408 (2016)

Caiani, A., Godin, A., Caverzasi, E., Gallegati, M., Kinsella, S., Stiglitz, J.E.: Agent based-stock flow consistent macroeconomics: Towards a benchmark model. Journal of Economic Dynamics and Control69, 375–408 (2016). https://doi.org/ 10.1016/j.jedc.2016.06.001

work page doi:10.1016/j.jedc.2016.06.001 2016
[10]

https://doi.org/10.1007/s40821-019-00121-0

Capone, G., Malerba, F., Nelson, R.R., Orsenigo, L., Winter, S.G.: History friendly models:retrospectiveandfutureperspectives.EurasianBusinessReview9(1),1–23 (March 2019). https://doi.org/10.1007/s40821-019-00121-0

work page doi:10.1007/s40821-019-00121-0 2019
[11]

MIT Press (2018), https://mitpress.mit.edu/books/ model-checking-second-edition

Clarke, E.M., Grumberg, O., Kroening, D., Peled, D.A., Veith, H.: Model checking, 2nd Edition. MIT Press (2018), https://mitpress.mit.edu/books/ model-checking-second-edition

work page 2018
[12]

In: Handbook of Compu- tational Economics, vol

Dawid, H., Gatti, D.D.: Agent-based macroeconomics. In: Handbook of Compu- tational Economics, vol. 4, pp. 63–156. Elsevier (2018). https://doi.org/10.2139/ ssrn.3112074

work page 2018
[13]

Journal of Economic Dy- namics and Control37(8), 1598–1625 (2013)

Dosi, G., Fagiolo, G., Napoletano, M., Roventini, A.: Income distribution, credit and fiscal policies in an agent-based Keynesian model. Journal of Economic Dy- namics and Control37(8), 1598–1625 (2013). https://doi.org/10.1016/j.jedc.2012. 11.008

work page doi:10.1016/j.jedc.2012 2013
[14]

Journal of Economic Dynamics and Control52, 166–189 (2015)

Dosi, G., Fagiolo, G., Napoletano, M., Roventini, A., Treibich, T.: Fiscal and mon- etary policies in complex evolving economies. Journal of Economic Dynamics and Control52, 166–189 (2015). https://doi.org/10.1016/j.jedc.2014.11.014

work page doi:10.1016/j.jedc.2014.11.014 2015
[15]

Journal of Economic Dynamics and Control34(9), 1748–1767 (2010)

Dosi, G., Fagiolo, G., Roventini, A.: Schumpeter meeting keynes: A policy-friendly model of endogenous growth and business cycles. Journal of Economic Dynamics and Control34(9), 1748–1767 (2010). https://doi.org/10.1016/j.jedc.2010.06.018 18 S. Blando et al

work page doi:10.1016/j.jedc.2010.06.018 2010
[16]

Economic Inquiry58(3), 1487–1516 (2020)

Dosi, G., Napoletano, M., Roventini, A., Stiglitz, J.E., Treibich, T.: Rational heuristics? expectations and behaviors in evolving economies with heterogeneous interacting agents. Economic Inquiry58(3), 1487–1516 (2020). https://doi.org/10. 1111/ecin.12897

work page 2020
[17]

Journal of Evolutionary Economics 27(1), 63–90 (2017)

Dosi, G., Napoletano, M., Roventini, A., Treibich, T.: Micro and macro policies in the keynes+schumpeter evolutionary models. Journal of Evolutionary Economics 27(1), 63–90 (2017). https://doi.org/10.1007/s00191-016-0466-4

work page doi:10.1007/s00191-016-0466-4 2017
[18]

https://doi.org/10.1007/s11403-019-00258-1, https://doi.org/ 10.1007/s11403-019-00258-1

Fagiolo, G., Giachini, D., Roventini, A.: Innovation, finance, and economic growth: anagent-basedapproach.JournalofEconomicInteractionandCoordination15(3), 703–736 (Jul 2020). https://doi.org/10.1007/s11403-019-00258-1, https://doi.org/ 10.1007/s11403-019-00258-1

work page doi:10.1007/s11403-019-00258-1 2020
[19]

In: Computer Simulation Validation, pp

Fagiolo, G., Guerini, M., Lamperti, F., Moneta, A., Roventini, A.: Validation of agent-based models in economics and finance. In: Computer Simulation Validation, pp. 763–787. Springer (2019). https://doi.org/10.1007/978-3-319-70766-2_31

work page doi:10.1007/978-3-319-70766-2_31 2019
[20]

Computational Economics30(3), 195–226 (2007)

Fagiolo, G., Moneta, A., Windrum, P.: A critical guide to empirical validation of agent-based models in economics: Methodologies, procedures, and open prob- lems. Computational Economics30(3), 195–226 (2007). https://doi.org/10.1007/ s10614-007-9104-4

work page 2007
[21]

Journal of Artificial Societies and Social Simulation20(1), 1 (2017)

Fagiolo, G., Roventini, A.: Macroeconomic policy in DSGE and agent-based models redux: New developments and challenges ahead. Journal of Artificial Societies and Social Simulation20(1), 1 (2017). https://doi.org/10.18564/jasss.3280

work page doi:10.18564/jasss.3280 2017
[22]

Financial Stability Board

Farmer, J.D., Foley, D.: The economy needs agent-based modelling. Nature 460(7256), 685–686 (2009). https://doi.org/10.1038/460685a

work page doi:10.1038/460685a 2009
[23]

In: Integrated Formal Methods - 13th In- ternational Conference, IFM 2017, Turin, Italy, September 20-22, 2017, Proceed- ings

Gilmore, S., Reijsbergen, D., Vandin, A.: Transient and steady-state statistical analysis for discrete event simulators. In: Integrated Formal Methods - 13th In- ternational Conference, IFM 2017, Turin, Italy, September 20-22, 2017, Proceed- ings. pp. 145–160 (2017). https://doi.org/10.1007/978-3-319-66845-1\_10, https: //doi.org/10.1007/978-3-319-66845-1_10

work page doi:10.1007/978-3-319-66845-1 2017
[24]

In: Albert, E., Sekerinski, E

Gilmore, S., Tribastone, M., Vandin, A.: An analysis pathway for the quantitative evaluation of public transport systems. In: Albert, E., Sekerinski, E. (eds.) Inte- grated Formal Methods - 11th International Conference, IFM 2014, Bertinoro, Italy, September 9-11, 2014, Proceedings. Lecture Notes in Computer Science, vol.8739,pp.71–86.Springer(2014).https...

work page doi:10.1007/978-3-319-10181-1_ 2014
[25]

Backpropagation through time and the brain.Current Opinion in Neurobiology, 55:82–89, 2019

Guerini, M., Moneta, A.: A method for agent-based models validation. Journal of Economic Dynamics and Control82, 125–141 (2017). https://doi.org/10.1016/j. jedc.2017.06.001

work page doi:10.1016/j 2017
[26]

Journal of Economic Dynamics and Control33(5), 1052–1072 (2009)

Heemeijer, P., Hommes, C., Sonnemans, J., Tuinstra, J.: Price stability and volatil- ity in markets with positive and negative expectations feedback: An experimen- tal investigation. Journal of Economic Dynamics and Control33(5), 1052–1072 (2009). https://doi.org/10.1016/j.jedc.2008.09.009, https://doi.org/10.1016/j.jedc. 2008.09.009

work page doi:10.1016/j.jedc.2008.09.009 2009
[27]

Journal of Economic Dynamics and Control35(1), 1–24 (2011)

Hommes, C.: The heterogeneous expectations hypothesis: Some evidence from the lab. Journal of Economic Dynamics and Control35(1), 1–24 (2011). https://doi. org/10.1016/j.jedc.2010.10.003, https://doi.org/10.1016/j.jedc.2010.10.003

work page doi:10.1016/j.jedc.2010.10.003 2011
[28]

In: Hand- book of Computational Economics, vol

Hommes, C.H.: Heterogeneous agent models in economics and finance. In: Hand- book of Computational Economics, vol. 2, pp. 1109–1186. Elsevier (2006). https: //doi.org/10.1016/s1574-0021(05)02023-x

work page doi:10.1016/s1574-0021(05)02023-x 2006
[29]

Routledge, London (2011)

Kirman, A.: Complex Economics: Individual and Collective Rationality. Routledge, London (2011). https://doi.org/10.23941/ejpe.v4i2.81 Statistical Model Checking of the K+S Model 19

work page doi:10.23941/ejpe.v4i2.81 2011
[30]

Journal of Economic Dynamics and Control90, 366–389 (2018)

Lamperti, F., Roventini, A., Sani, A.: Agent-based model calibration using ma- chine learning surrogates. Journal of Economic Dynamics and Control90, 366–389 (2018). https://doi.org/10.1016/j.jedc.2018.03.011

work page doi:10.1016/j.jedc.2018.03.011 2018
[31]

Journal of Artificial Societies and Social Simulation18(4), 4 (2015)

Lee, J.S., Filatova, T., Ligmann-Zielinska, A., Hassani-Mahmooei, B., Stonedahl, F., Lorscheid, I., Voinov, A., Polhill, J.G., Sun, Z., Parker, D.C.: The complexities of agent-based modeling output analysis. Journal of Artificial Societies and Social Simulation18(4), 4 (2015). https://doi.org/10.18564/jasss.2897

work page doi:10.18564/jasss.2897 2015
[32]

In: Runtime Verification (RV 2010)

Legay, A., Delahaye, B., Bensalem, S.: Statistical model checking: An overview. In: Runtime Verification (RV 2010). LNCS, vol. 6418, pp. 122–135. Springer (2010). https://doi.org/10.1007/978-3-642-16612-9_11

work page doi:10.1007/978-3-642-16612-9_11 2010
[33]

Harvard University Press, Cambridge, MA (1982)

Nelson, R.R., Winter, S.G.: An Evolutionary Theory of Economic Change. Harvard University Press, Cambridge, MA (1982). https://doi.org/10.1086/261177

work page doi:10.1086/261177 1982
[34]

Lamb, Jialin Yu, Philip H

Pangallo, M., Giachini, D., Vandin, A.: Statistical model checking of netlogo models. CoRRabs/2509.10977(2025). https://doi.org/10.48550/ARXIV.2509. 10977, https://doi.org/10.48550/arXiv.2509.10977

work page doi:10.48550/arxiv.2509 2025
[35]

In: International Conference on High Performance Computing & Simulation, HPCS 2014, Bologna, Italy, 21-25 July,

Pianini,D.,Sebastio,S.,Vandin,A.:Distributedstatisticalanalysisofcomplexsys- tems modeled through a chemical metaphor. In: International Conference on High Performance Computing & Simulation, HPCS 2014, Bologna, Italy, 21-25 July,

work page 2014
[36]

pp. 416–423. IEEE (2014). https://doi.org/10.1109/HPCSIM.2014.6903715, https://doi.org/10.1109/HPCSim.2014.6903715

work page doi:10.1109/hpcsim.2014.6903715 2014
[37]

Journal of Artificial Societies and Social Simulation 9(1), 15 (2006), https://www.jasss.org/9/1/15.html

Richiardi, M., Leombruni, R., Saam, N.J., Sonnessa, M.: A common protocol for agent-based social simulation. Journal of Artificial Societies and Social Simulation 9(1), 15 (2006), https://www.jasss.org/9/1/15.html

work page 2006
[38]

Performance Evaluation70(6), 457–475 (2013)

Sebastio, S., Vandin, A.: Multivesta: Statistical model checking for discrete event simulators. Performance Evaluation70(6), 457–475 (2013). https://doi.org/10. 4108/icst.valuetools.2013.254377

work page arXiv 2013
[39]

Computational and Math- ematical Organization Theory23(1), 94–121 (2017)

Secchi, D., Seri, R.: Controlling for false negatives in agent-based models: a review of power analysis in organizational research. Computational and Math- ematical Organization Theory23(1), 94–121 (2017). https://doi.org/10.1007/ s10588-016-9218-0, https://doi.org/10.1007/s10588-016-9218-0

work page doi:10.1007/s10588-016-9218-0 2017
[40]

The Quarterly Journal of Economics69(1), 99–118 (1955)

Simon, H.A.: A behavioral model of rational choice. The Quarterly Journal of Economics69(1), 99–118 (1955). https://doi.org/10.2307/1884852

work page doi:10.2307/1884852 1955
[41]

Tesfatsion, L., Judd, K.L.: Handbook of Computational Economics: Agent-Based Computational Economics, vol. 2. Elsevier, Amsterdam (2006). https://doi.org/ 10.1109/mci.2008.929849

work page doi:10.1109/mci.2008.929849 2006
[42]

Computational Economics60(4), 1507–1527 (2022)

Tieleman, S.: Towards a validation methodology for macroeconomic agent-based models. Computational Economics60(4), 1507–1527 (2022). https://doi.org/10. 1007/s10614-021-10191-w

work page 2022
[43]

Science185(4157), 1124–1131 (1974)

Tversky, A., Kahneman, D.: Judgment under uncertainty: Heuristics and biases. Science185(4157), 1124–1131 (1974). https://doi.org/10.1126/science.185.4157. 1124, https://doi.org/10.1126/science.185.4157.1124

work page doi:10.1126/science.185.4157 1974
[44]

In: Steffen, B

Vandin, A.: Statistical model checking of python agent-based models: An integra- tion of multivesta and mesa. In: Steffen, B. (ed.) Bridging the Gap Between AI and Reality - Second International Conference, AISoLA 2024, Crete, Greece, Octo- ber 30 - November 3, 2024, Proceedings. Lecture Notes in Computer Science, vol. 15217, pp. 398–419. Springer (2024)....

work page doi:10.1007/978-3-031-75434-0_ 2024
[45]

Journal of Economic Dynamics and Control143(10 2022)

Vandin, A., Giachini, D., Lamperti, F., Chiaromonte, F.: Automated and dis- tributed statistical analysis of economic agent-based models. Journal of Economic Dynamics and Control143(10 2022). https://doi.org/10.1016/j.jedc.2022.104458 20 S. Blando et al

work page doi:10.1016/j.jedc.2022.104458 2022
[46]

Sequential Tests of Statistical Hypotheses

Wald, A.: Sequential tests of statistical hypotheses. The Annals of Mathematical Statistics16(2), 117–186 (1945). https://doi.org/10.1214/aoms/1177731118

work page doi:10.1214/aoms/1177731118 1945
[47]

Wiley (2009), https: //books.google.it/books?id=X3ZQ7yeDn2IC

Wooldridge, M.: An Introduction to MultiAgent Systems. Wiley (2009), https: //books.google.it/books?id=X3ZQ7yeDn2IC

work page 2009
[48]

In: Computer Aided Verification (CAV 2002)

Younes, H.L.S., Simmons, R.G.: Probabilistic verification of discrete event systems using acceptance sampling. In: Computer Aided Verification (CAV 2002). LNCS, vol. 2404, pp. 223–235. Springer (2002) A Summary Tables This appendix collects the full observable-wise summary tables supporting the compactdiscussioninSection6.Tables4–6reporttheper-experiments...

work page 2002