{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,9,26]],"date-time":"2026-09-26T06:00:06Z","timestamp":1790402406258,"version":"4.1.0"},"posted":{"date-parts":[[2026]]},"group-title":"SSRN","reference-count":0,"publisher":"Elsevier BV","license":[{"start":{"date-parts":[[2026,1,1]],"date-time":"2026-01-01T00:00:00Z","timestamp":1767225600000},"content-version":"unspecified","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by-nc-nd\/4.0\/"}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":[],"abstract":"<jats:p>&lt;div&gt;\n &lt;p&gt;Purpose: This paper develops an integrated four-layer empirical monitoring framework for diagnosing multi-asset systemic stress, positioned within the empirical systemic-risk tradition of Kritzman et al. (2011) and Diebold and Yilmaz (2014) rather than as a structural equilibrium asset-pricing model. The objective is to provide institutional risk managers and macro analysts with a reproducible diagnostic architecture combining network topology, dynamic causality analysis, tail-risk quantification, and regime classification, applied concurrently with an active stress event. The paper does not aim to forecast returns, identify the stochastic discount factor, or test market efficiency.&lt;\/p&gt;\n &lt;p&gt;Design\/methodology\/approach: Four methodological layers are integrated: network topology via Mantegna (1999) correlation distance, multidimensional scaling, Ward clustering, and minimum spanning trees; dynamic causality via pairwise Granger testing with Bonferroni correction, lead-lag cross-correlation, and joint F-tests in a nine-asset VAR; tail risk via Conditional Value-at-Risk and Longin and Solnik (2001) tail dependence; regime classification via the Kritzman et al. (2011) Absorption Ratio and a hidden Markov model with BIC selection and out-of-sample evaluation. Analysis covers nine global asset classes from April 2025 to April 2026 (N = 261), supplemented by a 14-year long-history sample (N = 3,472).&lt;\/p&gt;\n &lt;p&gt;Findings: The Absorption Ratio 0.50 threshold (Kritzman et al., 2011) is breached in all four crisis windows tested (the Global Financial Crisis, COVID-19, the 2022\u20132023 Federal Reserve tightening cycle, and the current-study window), providing multi-episode threshold validation. Bonferroni correction reduces 15 uncorrected Granger relationships to 4, all pointing to Japan Equity. The DCC-GARCH minimum spanning tree converges with the Granger evidence on Japan Equity centrality. The out-of-sample hidden Markov model, trained on 14 years of multi-crisis data and applied without re-estimation, correctly identifies the April 2026 stress regime. Joint F-tests support Oil quasi-exogeneity in conditional mean while variance decomposition reveals volatility-channel transmission to equities and rates.&lt;\/p&gt;\n &lt;p&gt;Research limitations\/implications: The 261-observation current-study window is short for systemic-risk modelling; conclusions specific to this window are exploratory. The multi-crisis Absorption Ratio replication, long-history DCC-GARCH baseline, and out-of-sample hidden Markov model evaluation mitigate but do not eliminate single-episode concerns. Granger findings are subject to heteroskedasticity, non-linear dependence, and finite-sample size distortions; only Bonferroni-surviving results are treated as robust. The framework provides no equilibrium interpretation. Future research should pursue walk-forward hidden Markov model evaluation on larger samples, asymmetric and block DCC specifications, Student-t and copula-based tail dependence extensions, and transfer-entropy or non-linear Granger frameworks for non-linear predictability structures.&lt;\/p&gt;\n &lt;p&gt;Practical implications: The framework offers institutional risk managers a reproducible monitoring architecture combining four methodologically independent diagnostics. Portfolio managers obtain a multi-method check on diversification assumptions during stress periods. The volume-weighted minimum spanning tree provides a capital-commitment-aware view of cross-asset structure that complements the price-correlation view. The out-of-sample hidden Markov model enables real-time regime classification using only historical data. Risk officers receive defensible directional findings (Bonferroni-corrected Granger) that survive the most stringent multiple-testing correction. Macro strategists gain a transparent map of which assets transmit shocks through mean predictability versus through volatility-channel effects on forecast-error variance.&lt;\/p&gt;\n &lt;p&gt;Social implications: Open and reproducible systemic-risk monitoring tools support broader financial-stability objectives valued by regulators, central banks, and institutional fiduciaries who serve household savers, pension beneficiaries, and insurance policyholders. The framework's open-source code and fully bundled data pipeline lower the methodological barrier for smaller institutions, academic researchers in emerging markets, and graduate students who may lack access to proprietary risk platforms. By providing methodologically triangulated diagnostics rather than single-method signals, the framework helps reduce overconfidence in any one indicator, a recurring contributor to retail and institutional losses during prior systemic crises.&lt;\/p&gt;\n &lt;p&gt;&lt;br&gt;&lt;\/p&gt;\n&lt;\/div&gt;<\/jats:p>","DOI":"10.2139\/ssrn.7521018","type":"posted-content","created":{"date-parts":[[2026,9,26]],"date-time":"2026-09-26T05:20:13Z","timestamp":1790400013000},"source":"Crossref","is-referenced-by-count":0,"title":["The Geometry of Risk: An Integrated Monitoring Framework for Multi-Asset Systemic Stress"],"prefix":"10.2139","author":[{"ORCID":"https:\/\/orcid.org\/0009-0009-5863-2952","authenticated-orcid":true,"given":"Youness","family":"Yachruti","sequence":"first","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"78","container-title":[],"original-title":[],"deposited":{"date-parts":[[2026,9,26]],"date-time":"2026-09-26T05:20:13Z","timestamp":1790400013000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.ssrn.com\/abstract=7521018"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026]]},"references-count":0,"URL":"https:\/\/doi.org\/10.2139\/ssrn.7521018","relation":{},"subject":[],"published":{"date-parts":[[2026]]},"subtype":"preprint"}}