{"schema_version":1,"experiment_id":"eoa_inverse_innovation_exp11_mechanism_context_external20_20260804","research_id":"eoa_inverse_innovation_exp11_external_scrutiny_20260804","cell_id":"invariant_mode_decomposition_design__economics_finance","opaque_id":"invariant_mode_decomposition_design__economics_finance__A","search_lanes":{"direct_problem":{"queries":["\"dynamic mode decomposition\" finance financial markets risk","bank balance sheet transition matrix eigenvalue stress testing deleveraging spiral","The interaction of bank leverage interest rate risk and runnable funding"],"source_ids":["SRC1","SRC2","SRC3"],"no_result_note":null},"closest_prior_art":{"queries":["site:bankofengland.co.uk RAMSI balance sheet feedback stress testing model","systemic risk eigenvalue bank balance sheet stability matrix early warning paper","dynamic mode decomposition with control journal paper official DOI"],"source_ids":["SRC4","SRC5","SRC6"],"no_result_note":null},"historical_terminology":{"queries":["financial instability eigenvalue early warning bank leverage liquidity duration funding","Brunnermeier Pedersen Market Liquidity and Funding Liquidity PDF 2009 official","debt deflation leverage cycle liquidity spiral bank balance sheet"],"source_ids":["SRC2","SRC5"],"no_result_note":null},"products_practices_standards":{"queries":["site:bis.org bank integrated balance sheet stress testing feedback effects liquidity solvency","site:bankofengland.co.uk RAMSI balance sheet feedback stress testing model","Basel RMA30 stress testing feedback second-round effects reverse stress test","bank stress testing different balance sheet assumptions Bundesbank 2017 abstract"],"source_ids":["SRC3","SRC4","SRC7","SRC8"],"no_result_note":null},"non_english_regional":{"queries":["Bank Bilanz Stress Eigenwert Dynamik Liquidität Solvenz Gesamtbanksteuerung","\"动态模态分解\" 银行 风险 资产负债 流动性","\"mode dynamique\" bilan bancaire stress liquidité solvabilité"],"source_ids":["SRC8"],"no_result_note":null},"composition_subproblems":{"queries":["bank balance sheet transition matrix eigenvalue stress testing deleveraging spiral","financial instability eigenvalue early warning bank leverage liquidity duration funding","dynamic mode decomposition with control journal paper official DOI","bank stress testing different balance sheet assumptions Bundesbank 2017 abstract"],"source_ids":["SRC1","SRC3","SRC5","SRC6","SRC8"],"no_result_note":null}},"sources":[{"source_id":"SRC1","title":"The interaction of bank leverage, interest rate risk, and runnable funding","url":"https://www.federalreserve.gov/econres/notes/feds-notes/the-interaction-of-bank-leverage-interest-rate-risk-and-runnable-funding-20240830.html","publisher":"Board of Governors of the Federal Reserve System","date_or_year":"2024-08-30","source_type":"PRIMARY_RESEARCH","language":"English","claims_supported":["Capital, usable liquidity, asset duration and runnable funding interact in ways that materially affected recent bank failures.","A joint liquidity-solvency measure identified vulnerable banks whose interaction of weaknesses had been overlooked by markets.","Duration-related fair-value losses can make nominally liquid assets unusable without breaching capital requirements.","Contingent liquidity readiness or additional capital are identifiable response classes, but their suitability depends on the institution's joint vulnerability."]},{"source_id":"SRC2","title":"Market Liquidity and Funding Liquidity","url":"https://pages.stern.nyu.edu/~lpederse/papers/Mkt_Fun_Liquidity.pdf","publisher":"The Review of Financial Studies / New York University Stern School of Business","date_or_year":"2009","source_type":"PRIMARY_RESEARCH","language":"English","claims_supported":["Funding constraints, margins, losses and market liquidity can form self-reinforcing loss and margin spirals.","When capital constraints bind, leveraged institutions reduce positions and market liquidity deteriorates.","Small losses can cause discontinuous liquidity deterioration under fragile, multiple-equilibrium conditions.","This establishes an older causal vocabulary—liquidity spirals, deleveraging and destabilizing margins—for the proposed coupled-stress problem."]},{"source_id":"SRC3","title":"Making supervisory stress tests more macroprudential: Considering liquidity and solvency interactions and systemic risk","url":"https://www.bis.org/bcbs/publ/wp29.htm","publisher":"Basel Committee on Banking Supervision, Bank for International Settlements","date_or_year":"2015","source_type":"OFFICIAL_GUIDANCE","language":"English","claims_supported":["Stand-alone treatment of liquidity and solvency can underestimate financial-system vulnerabilities.","Integrated liquidity-solvency stress testing is an officially documented supervisory direction.","Micro-level interaction estimates, network analysis and agent-based models are recognized ways to incorporate feedback and endogenous bank behavior.","The source establishes substantial adjacent practice but does not document institution-level eigenmode detection or robust modal damping."]},{"source_id":"SRC4","title":"RAMSI: a top-down stress-testing model","url":"https://www.bankofengland.co.uk/financial-stability-paper/2012/ramsi-a-top-down-stress-testing-model","publisher":"Bank of England","date_or_year":"2012-09-11","source_type":"FIRST_PARTY_PRODUCT","language":"English","claims_supported":["RAMSI is an implemented top-down model in the Bank of England's risk-assessment toolkit.","The model assesses multiple banking-system vulnerabilities and was used in the 2011 UK Financial Sector Assessment Program.","It is close prior art for dynamic balance-sheet stress modeling and feedback analysis, though not for the proposed local invariant-mode and worst-case-gain rule."]},{"source_id":"SRC5","title":"Early warning of systemic risk in global banking: eigen-pair R number for financial contagion and market price-based methods","url":"https://repository.essex.ac.uk/30689/","publisher":"Annals of Operations Research / University of Essex Research Repository","date_or_year":"2021 online; 2023 volume","source_type":"PRIMARY_RESEARCH","language":"English","claims_supported":["A spectral eigen-pair method has already been applied to bilateral bank balance-sheet data as a dynamical-system stability problem.","The reported dominant eigen-pair supplies an early-warning tipping-point measure and identifies vulnerable banking systems.","This is the closest same-lever finance analogue, but it operates on a cross-border exposure network rather than one institution's weekly multivariate transition.","It does not document action-conditioned modal damping with held-out and affected-party criteria."]},{"source_id":"SRC6","title":"Dynamic Mode Decomposition with Control","url":"https://epubs.siam.org/doi/10.1137/15M1013857","publisher":"Society for Industrial and Applied Mathematics","date_or_year":"2016-01-26","source_type":"PRIMARY_RESEARCH","language":"English","claims_supported":["Dynamic mode decomposition with control already extracts low-order dynamics and actuation effects from time-ordered observations.","Ordinary DMD cannot by itself distinguish autonomous dynamics from external forcing.","The general mathematical causal lever underlying action-conditioned transition maps is established prior art.","The demonstrations are outside bank balance-sheet risk and do not validate causal identification from sparse financial action episodes."]},{"source_id":"SRC7","title":"RMA30 — Stress testing","url":"https://www.bis.org/basel_consolidated_guidelines/chapter/RMA/30.htm","publisher":"Basel Committee on Banking Supervision, Bank for International Settlements","date_or_year":"2026 consolidated framework","source_type":"OFFICIAL_STANDARD","language":"English","claims_supported":["Bank stress testing is an established supervisory and risk-management practice spanning enterprise-wide and reverse stress tests.","Risk identification should cover on- and off-balance-sheet exposures and factors affecting both solvency and liquidity.","Supervisory stress tests may include feedback and second-round effects and must be used with other analytical or policy instruments.","The standard supports identifiable institutional and supervisory authorizers but does not prescribe invariant-mode monitoring."]},{"source_id":"SRC8","title":"Bank stress testing under different balance sheet assumptions","url":"https://www.bundesbank.de/resource/blob/704084/9bf00fa01a9fb3db9c32b6596fd146dc/mL/2017-04-10-dkp-07-data.pdf","publisher":"Deutsche Bundesbank","date_or_year":"2017","source_type":"PRIMARY_RESEARCH","language":"English with German front matter and German banking context","claims_supported":["German supervisory survey data show that dynamic balance-sheet adjustments become material over longer stress horizons.","Bank and depositor behavior changes balance-sheet size and composition after interest-rate shocks.","Individually rational balance-sheet contraction can reduce aggregate credit supply when many banks react similarly.","Constrained dynamic balance sheets are already part of regional stress-testing terminology and practice, but the study does not use spectral modes or modal control."]}],"problem_evidence":{"status":"PARTLY_SUPPORTED","finding":"The underlying problem exists: official and primary sources show that leverage, usable liquidity, duration, runnable funding, solvency and market liquidity interact, and that analyzing these risks separately can understate vulnerability or miss amplification. The evidence does not directly establish the proposal's stronger formulation that all named coordinate limits remain satisfied while a locally estimated invariant mode reliably approaches instability.","source_ids":["SRC1","SRC2","SRC3","SRC7","SRC8"],"uncertainty":"Published evidence supports coupled fragility and late recognition, but not the existence, persistence or incremental predictive value of the proposed weekly eigenmodes for a particular institution."},"adopter_evidence":{"status":"SUPPORTED","finding":"Institutional treasury and risk teams, risk committees, senior management and prudential supervisors are identifiable users and authorizers. Official standards assign stress-testing responsibilities to banks and authorities, while RAMSI demonstrates use by a central-bank financial-stability function.","source_ids":["SRC3","SRC4","SRC7","SRC8"],"uncertainty":"The precise approval chain for live balance-sheet changes is jurisdiction- and institution-specific; supervisory non-objection or approval may be required."},"implementation_evidence":{"status":"PARTLY_SUPPORTED","finding":"All major technical ingredients exist separately: dynamic balance-sheet stress models, integrated liquidity-solvency analysis, spectral balance-sheet early warning and DMD with control. No retained source validates their proposed composition on one institution's weekly state, demonstrates reliable mode separation under financial regime change, or identifies safe damping effects from sparse action episodes.","source_ids":["SRC3","SRC4","SRC5","SRC6","SRC8"],"uncertainty":"Weekly sample size, structural breaks, reflexivity, measurement error, rare exogenous actions and correlated interventions may prevent stable mode estimation or credible action-response identification."},"prior_art":{"disposition":"ADJACENT_PRIOR_ART","closest_analogues":[{"name":"Spectral eigen-pair early warning on global-bank balance-sheet networks","source_ids":["SRC5"],"same_problem":true,"same_causal_lever":true,"overlap":"Uses a dominant eigen-pair of a balance-sheet-based stability matrix to detect an approaching systemic tipping point and outperform contemporaneous market measures.","remaining_difference":"It analyzes cross-border contagion among banking systems, not a local weekly transition for one institution, and does not test action-conditioned worst-case modal damping or protected-party incidence."},{"name":"Integrated liquidity-solvency stress testing and RAMSI","source_ids":["SRC3","SRC4","SRC7"],"same_problem":true,"same_causal_lever":false,"overlap":"Models interacting balance-sheet risks, dynamic behavior, feedback and second-round effects for supervisory stress assessment.","remaining_difference":"These approaches are scenario, accounting, network or simulation based rather than selecting persistent local eigenmodes and ranking actions by robust reductions in modal gain."},{"name":"Dynamic mode decomposition with control","source_ids":["SRC6"],"same_problem":false,"same_causal_lever":true,"overlap":"Estimates low-order modes and separates actuation effects from underlying dynamics using observed state and control snapshots.","remaining_difference":"It is a general control method demonstrated outside bank risk; it supplies neither financial identification assumptions nor evidence that permissible balance-sheet actions damp a deleveraging mode."},{"name":"Dynamic balance-sheet stress testing under behavioral adjustment","source_ids":["SRC8"],"same_problem":true,"same_causal_lever":false,"overlap":"Uses supervisory data to distinguish static price effects from dynamic quantity adjustments and recognizes credit-supply externalities from common contraction.","remaining_difference":"It estimates scenario-specific balance-sheet reactions over annual horizons rather than weekly invariant directions, spectral gains or robust action-conditioned mode changes."}],"contrastive_claim_remaining":"For a prespecified institution and regime, a regularized weekly transition-mode detector adds locked-holdout stress-detection value beyond named limits and a documented transition-based rival; separately, for an action class with identified quasi-experimental response or an explicit admissible response set, the smallest permitted action lowers the selected mode's worst-case gain relative to no action and the rival without violating residual, drift or affected-party budgets.","contrastive_claim_falsifier":"The incremental claim is falsified if selected modes fail prespecified holdout detection, separation, residual or drift tests; or, after detection passes, no action response is identified or bounded, the admissible set contains no gain reduction, or held-out action episodes fail improvement and incidence criteria against the rival.","confidence":"MODERATE","search_limitations":"The bounded search used eight retained sources across six lanes. It did not exhaust patents, proprietary bank models, vendor documentation, confidential supervisory methods, all jurisdictions or all non-English databases. Component-wise prior art may exist in undisclosed implementations."},"researchability_gates":{"externally_supported_problem":{"status":"PASS","rationale":"Multiple independent official and primary sources document coupled liquidity, solvency, leverage, duration, funding and fire-sale dynamics and the shortcomings of isolated analysis.","source_ids":["SRC1","SRC2","SRC3","SRC7","SRC8"]},"identifiable_adopter_or_authorizer":{"status":"PASS","rationale":"Bank risk functions, risk committees, management and prudential supervisors are identifiable; official standards and RAMSI demonstrate corresponding institutional roles.","source_ids":["SRC3","SRC4","SRC7"]},"distinct_testable_incremental_claim":{"status":"PASS","rationale":"Despite adjacent spectral and stress-testing art, the institution-level locked-holdout comparison plus separately identified or bounded worst-case modal damping remains distinct and falsifiable.","source_ids":["SRC3","SRC4","SRC5","SRC6","SRC8"]},"bounded_next_evidence_step":{"status":"PASS","rationale":"A preregistered no-action retrospective study using the specified minimum observations, data splits, fixed thresholds, ordinary-limit baseline and transition rival is bounded and can reject detection before any live intervention.","source_ids":["SRC1","SRC3","SRC5","SRC6","SRC8"]},"no_unresolved_safety_or_authority_stop":{"status":"PASS","rationale":"The first step is analysis only; live trading, asset sales, credit restrictions and policy changes are excluded. Management and supervisors retain live-action authority, and failure of identification, residual, drift or incidence criteria triggers no action.","source_ids":["SRC2","SRC3","SRC7","SRC8"]},"adequate_search_evidence":{"status":"PASS","rationale":"The bounded adversarial search covered all six required lanes and retained exactly eight directly relevant sources from seven publisher contexts, including official standards, official institutional research, first-party model documentation and peer-reviewed primary research.","source_ids":["SRC1","SRC2","SRC3","SRC4","SRC5","SRC6","SRC7","SRC8"]}},"strict_success":true,"screen_survival":true,"remaining_research_value":"HIGH","recommended_next_step":"Conduct only the preregistered retrospective no-action study. First document a concrete transition-based rival and lock detection, separation, residual, drift and incidence tests. Advance to an action-specific shadow analysis only for an action class supported by at least 12 plausibly exogenous episodes or a defensible accounting-based admissible response set; do not recommend live damping from the observational transition alone.","world_novelty_boundary":"This bounded search supports an adjacent-prior-art disposition and a remaining empirical contrast, not world novelty. It cannot establish patentability, freedom to operate, absence of proprietary implementations, market size, regulatory acceptance or realized impact."}