[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-detail-84500-en":3,"doc-seo-84500-105":29,"detail-sidebar-cat-0-en-105":91},{"code":4,"msg":5,"data":6},0,"success",{"doc_id":7,"user_id":8,"nickname":9,"user_avatar":10,"doc_module":4,"category_id":11,"category_name":12,"doc_title":13,"doc_description":14,"doc_content":15,"file_id":16,"file_url":17,"file_type":18,"file_size":19,"view_count":20,"is_deleted":4,"is_public":20,"is_downloadable":20,"audit_status":20,"page_count":21,"language":22,"language_code":23,"site_id":24,"html_lang":23,"table_of_contents":25,"faqs":26,"seo_title":13,"seo_description":14,"update_tm":27,"read_time":28},84500,549758146520,"Patrick","https://ap-avatar.wpscdn.com/avatar/80002397d8c0411e94?_k=1775819394049821470",8,"Research & Report","Recursive Multi-Agent Trading System Iterative Optimized Portfolio Strategy Under Geopolitical Uncertainty","Geopolitical risk creates persistent market stress that conventional quantitative models struggle to incorporate in real time. The RMATS framework proposes a Recursive Multi-Agent Trading System that coordinates four specialist agents—Sentiment, Report, Analysis, and Risk—via a formally specified recursive protocol plus a GPR-triggered circuit breaker. When the Caldara–Iacoviello GPR index exceeds the 60th percentile of 2021–2022 calibration, the Risk Agent shifts the portfolio toward defensive assets. Over 812 trading days, RMATS attains the highest event-period Sharpe (1.571) and lowest MDD (7.52%) among tested strategies, with circuit-breaker ablation reducing event-window MDD by 0.84 percentage points.","Recursive Multi-Agent Trading System: Iterative Optimized Portfolio Strategy Under Geopolitical Uncertainty  \n* Jing Yang  \n[jing.y@wustl.edu](jing.y@wustl.edu)[ ](jing.y@wustl.edu)Washington University in St. Louis St. Louis, MO, USA  \nPenghao Liang  \n[liang.p@northeastern.edu](liang.p@northeastern.edu)[ ](liang.p@northeastern.edu)Northeastern University Boston, MA, USA  \nYichao Wu  \n[wu.yicha@northeastern.edu](wu.yicha@northeastern.edu)[ ](wu.yicha@northeastern.edu)Northeastern University Boston, MA, USA  \nMengwei Yuan  \n[yuanmw1998@gmail.com](yuanmw1998@gmail.com)[ ](yuanmw1998@gmail.com)Independent Researcher Milpitas, CA, USA  \nJianan Liu  \n[jiananliu2408@gmail.com](jiananliu2408@gmail.com)[ ](jiananliu2408@gmail.com)Independent Researcher Austin, TX, USA  \nXianyou Li  \n[xl4230@nyu.edu](xl4230@nyu.edu)[ ](xl4230@nyu.edu)New York University New York, NY, USA  \narXiv :2605 .25311v2 [ cs .MA] 12 Jul 2026  \nWeiran Yan  \n[yanwr2016@gmail.com](yanwr2016@gmail.com)[ ](yanwr2016@gmail.com)Independent Researcher Santa Clara, CA, USA  \nJiayi Yang  \n[yangjiay@northeastern.edu](yangjiay@northeastern.edu)[ ](yangjiay@northeastern.edu)Northeastern University Boston, MA, USA  \nAbstract  \nGeopolitical risk constitutes a distinct and persistent source of market stress that traditional quantitative frameworks struggle to incorporate in real time. This paper proposes RMATS, a Recursive Multi-Agent Trading System that integrates four specialist agents (Sentiment, Report, Analysis, Risk) through a formally specified recursive coordination protocol and a GPR-triggered circuit breaker. When the Caldara–Iacoviello geopolitical risk index exceeds the 60th percentile of its 2021–2022 calibration distribution, the Risk Agent hard-shifts the portfolio toward defensive assets. Over 812 trading days (January 2023–March 2026), RMATS achieves the highest event-period Sharpe ratio (1.571) and lowest MDD (7.52%) among all strategies during 300 geopolitical stress event days, outperforming MAPS-lite (1.346) and the GPR rule-based baseline (0.981) . Circuit breaker ablation confirms a +0.84pp MDD reduction in event windows versus the no-CB variant. Full-period Sharpe is 0.953, outperforming MAPS-lite (0.836, Δ = +0 . 117), with strong non-event performance (Sharpe +0.543 over 512 non-event days) . The recursive protocol (median 1 round, 97.4% within 􀁁 ≤2, max 3, 39 steps) currently functions as an architectural framework for future heterogeneous signal integration; the 􀁁 =1 ablation (ΔSharpe = −0 . 002) confirms its performance contribution is negligible under the current monthly-GPR configuration. All results are verified with real yfinance data and the Caldara–Iacoviello GPR index [25] .  \nCCS Concepts  \n• Computing methodologies → Reinforcement learning; Multiagent systems; Natural language processing.  \nKeywords  \nmulti-agent systems, portfolio optimization, geopolitical risk, GPR index, reinforcement learning, drawdown control  \n1 Introduction  \nFinancial markets face growing exposure to geopolitical shocks that traditional statistical models are ill-equipped to handle in real time.  \nCrises ranging from military conflicts and trade disputes to energy supply disruptions and banking failures routinely trigger crossasset dislocations that overwhelm rule-based and pattern-based quantitative strategies [1, 2] . Existing frameworks, built on the assumption of relatively stable data-generating processes, often lack the adaptive mechanisms needed to respond to abrupt geopolitical structural breaks [3] .  \nPrior work on multi-agent portfolio systems, including MAPS [16] and MSPM [17], has demonstrated that task specialisation across agents enhances risk-adjusted returns. Yet these systems share a common limitation: they rely on fixed, single-pass pipelines with no iterative inter-agent revision, offer no formal guarantees of weight convergence, and none incorporates real-time geopolitical risk sensing as a first-class capital preservation mechanism.  \nThis pap","cbCaijHSZQjHe0JZ","https://ap.wps.com/l/cbCaijHSZQjHe0JZ","pdf",1040625,1,9,"English","en",105,"# Introduction\n## Related Work\n### Financial Large Language Models and Geopolitical Text Analysis\n# Methodology\n## Recursive coordination protocol and GPR-triggered circuit breaker\n## Specialist agents (Sentiment, Report, Analysis, Risk)\n# Experimental Evaluation\n## Event-period and non-event performance\n## Ablation studies and limitations","[{\"question\":\"What problem does RMATS address in quantitative trading under geopolitical uncertainty?\",\"answer\":\"RMATS targets the difficulty traditional quantitative models have in incorporating geopolitical stress signals in real time, especially during abrupt structural breaks that disrupt cross-asset relationships.\"},{\"question\":\"How does RMATS use the GPR index and the circuit breaker?\",\"answer\":\"RMATS monitors the Caldara–Iacoviello geopolitical risk (GPR) index and, once it exceeds the 60th percentile of its 2021–2022 calibration distribution, the Risk Agent hard-shifts the portfolio toward defensive assets via a circuit breaker mechanism.\"},{\"question\":\"What evidence supports RMATS’s performance and the role of the circuit breaker and recursion protocol?\",\"answer\":\"Over 812 trading days, RMATS achieves the highest event-period Sharpe (1.571) and lowest MDD (7.52%). Circuit breaker ablation shows a +0.84pp MDD reduction in event windows versus no circuit breaker, while recursion protocol ablation reports negligible measurable performance gains under the current monthly-GPR configuration (ΔSharpe = −0.002).\"}]",1784196089,23,{"code":4,"msg":30,"data":31},"ok",{"site_id":24,"language":23,"slug":32,"title":13,"keywords":33,"description":14,"schema_data":34,"social_meta":86,"head_meta":88,"extra_data":90,"updated_unix":27},"recursive-multi-agent-trading-system-iterative-optimized-portfolio-strategy-under-geopolitical-uncertainty","",{"@graph":35,"@context":85},[36,53,68],{"@type":37,"itemListElement":38},"BreadcrumbList",[39,43,47,50],{"item":40,"name":41,"@type":42,"position":20},"https://docshare.wps.com","Home","ListItem",{"item":44,"name":45,"@type":42,"position":46},"https://docshare.wps.com/document/","Document",2,{"item":48,"name":12,"@type":42,"position":49},"https://docshare.wps.com/document/research-report/",3,{"item":51,"name":13,"@type":42,"position":52},"https://docshare.wps.com/document/recursive-multi-agent-trading-system-iterative-optimized-portfolio-strategy-under-geopolitical-uncertainty/84500/",4,{"url":51,"name":13,"@type":54,"author":55,"headline":13,"publisher":57,"fileFormat":60,"inLanguage":23,"description":14,"dateModified":61,"datePublished":62,"encodingFormat":60,"isAccessibleForFree":63,"interactionStatistic":64},"DigitalDocument",{"name":9,"@type":56},"Person",{"url":40,"name":58,"@type":59},"DocShare","Organization","application/pdf","2026-07-17","2026-07-16",true,{"@type":65,"interactionType":66,"userInteractionCount":20},"InteractionCounter",{"@type":67},"ViewAction",{"@type":69,"mainEntity":70},"FAQPage",[71,77,81],{"name":72,"@type":73,"acceptedAnswer":74},"What problem does RMATS address in quantitative trading under geopolitical uncertainty?","Question",{"text":75,"@type":76},"RMATS targets the difficulty traditional quantitative models have in incorporating geopolitical stress signals in real time, especially during abrupt structural breaks that disrupt cross-asset relationships.","Answer",{"name":78,"@type":73,"acceptedAnswer":79},"How does RMATS use the GPR index and the circuit breaker?",{"text":80,"@type":76},"RMATS monitors the Caldara–Iacoviello geopolitical risk (GPR) index and, once it exceeds the 60th percentile of its 2021–2022 calibration distribution, the Risk Agent hard-shifts the portfolio toward defensive assets via a circuit breaker mechanism.",{"name":82,"@type":73,"acceptedAnswer":83},"What evidence supports RMATS’s performance and the role of the circuit breaker and recursion protocol?",{"text":84,"@type":76},"Over 812 trading days, RMATS achieves the highest event-period Sharpe (1.571) and lowest MDD (7.52%). Circuit breaker ablation shows a +0.84pp MDD reduction in event windows versus no circuit breaker, while recursion protocol ablation reports negligible measurable performance gains under the current monthly-GPR configuration (ΔSharpe = −0.002).","https://schema.org",{"og:url":51,"og:type":87,"og:title":13,"og:site_name":58,"og:description":14},"article",{"robots":89,"canonical":51},"index,follow",{"doc_id":7,"site_id":24},{"code":4,"msg":5,"data":92},[93,97,101,105,110,115,120,123,127,130,134],{"id":20,"doc_module":4,"doc_module_name":45,"category_name":94,"show_sort_weight":95,"slug":96},"Story & Novel",90,"story-novel",{"id":46,"doc_module":4,"doc_module_name":45,"category_name":98,"show_sort_weight":99,"slug":100},"Literature",80,"literature",{"id":52,"doc_module":4,"doc_module_name":45,"category_name":102,"show_sort_weight":103,"slug":104},"Exam",70,"exam",{"id":106,"doc_module":4,"doc_module_name":45,"category_name":107,"show_sort_weight":108,"slug":109},5,"Comic",60,"comic",{"id":111,"doc_module":4,"doc_module_name":45,"category_name":112,"show_sort_weight":113,"slug":114},6,"Technology",50,"technology",{"id":116,"doc_module":4,"doc_module_name":45,"category_name":117,"show_sort_weight":118,"slug":119},7,"Healthcare",40,"healthcare",{"id":11,"doc_module":4,"doc_module_name":45,"category_name":12,"show_sort_weight":121,"slug":122},30,"research-report",{"id":21,"doc_module":4,"doc_module_name":45,"category_name":124,"show_sort_weight":125,"slug":126},"Religion & Spirituality",20,"religion-spirituality",{"id":125,"doc_module":4,"doc_module_name":45,"category_name":128,"show_sort_weight":125,"slug":129},"World Cup","world-cup",{"id":131,"doc_module":4,"doc_module_name":45,"category_name":132,"show_sort_weight":131,"slug":133},10,"Lifestyle","lifestyle",{"id":135,"doc_module":4,"doc_module_name":45,"category_name":136,"show_sort_weight":106,"slug":137},19,"General","general"]