{"paper_id":"2408.12839","title":"Causal Hierarchy in the Financial Market Network -- Uncovered by the Helmholtz-Hodge-Kodaira Decomposition","authors":["Tobias Wand","Oliver Kamps","Hiroshi Iyetomi"],"paper_date":"2024-08-23","paper_type":"empirical","paper_url":"https://arxiv.org/pdf/2408.12839v1","page_url":"https://thequant.space/flowcharts/causal-hierarchy-in-the-financial-market-network----uncovere/","scores":{"empirical_rigor":7,"math_complexity":8.5,"quadrant":"Holy Grail"},"proposes":{"status":"available","statement":"The paper applies the Helmholtz-Hodge-Kodaira decomposition to Granger causality networks to identify causal drivers and hierarchical flow in financial sector interactions.","quote":"Using the HHKD on this Granger network then reveals if a sector is rather driven by other sectors or if it is a causal driver of the whole system. First, this article will present the database, the algorithm from [8] arXiv:2408.12839v1 [q-fin.ST] 23 Aug 2024 Wand et al., Causal Hierarchy in the Financial Market Network to estimate Granger causality networks and the HHKD for graphs in section 2.","summary":"Granger causality can uncover the cause and effect relationships in financial networks. However, such networks can be convoluted and difficult to interpret, but the Helmholtz-Hodge-Kodaira decomposition can split them into a rotational and gradient component which reveals the hierarchy of Granger…","strategy":null},"assumptions":{"status":"available","items":[{"assumption":"Returns are calculated as daily price changes, ensuring stationarity of the data.","type":"data","stated":"explicit","quote":"The calculation of returns ensures the stationarity of the data and this database is updated continuously with further details on the data curation given in [23].","support":"full"},{"assumption":"Sector portfolios are constructed as value‑weighted averages of all stocks in each sector listed on NYSE, AMEX and NASDAQ.","type":"data","stated":"explicit","quote":"These portfolios are constructed as value-weighted averages of all stocks in a business sector listed on NYSE, AMEX and NASDAQ and the data consists of the daily returns R(i) t = \u0010 P (i) t −P (i) t−1 \u0011 /P (i) t−1 of these portfolios’ prices.","support":"full"},{"assumption":"The Restricted Conditional Granger Causality Index (RCGCI) provides a sparse full model to avoid over‑fitting.","type":"model","stated":"explicit","quote":"A comparative study of multivariate Granger networks [9] indicates that the Restricted conditional Granger causality index (RCGCI) from [8] is the most suitable Granger causality estimation scheme for the financial data analysed in this article. Hence, the resulting full model may not contain lagged representations of all possible explanatory variables but only some selected \u0000X(i1), . . . , X(iI)\u0001 and therefore","support":"full"},{"assumption":"The decomposition assumes the network is sufficiently closed so that background flux can be ignored.","type":"model","stated":"implied","quote":"Note that a third component can exist and represents a background flux into and out of the system, but is usually ignored as one assumes that the system of interest is sufficiently closed.","support":"full"},{"assumption":"Synthetic cyclic networks used for validation are representative of real financial networks.","type":"other","stated":"implied","quote":"Though not shown here, for synthetic cyclic networks, the RCGCI also successfully estimates the loop topology.","support":"full"}]},"datasets":{"status":"available","items":[{"name":"Kenneth French business sector return time series","provider":"Kenneth French","kind":"market-prices","access":"public","period":null,"universe":"49 business sectors","quote":"2 Materials and Methods 2.1 Data To analyse the interactions between different sectors of the economy, we use the database of Ken French which contains the return time series of representative portfolios for 49 different business sectors [22].","support":"full"}],"known":[],"sample":{"start":null,"end":null,"market":null,"asset_class":null,"frequency":null},"data_resolution":"daily"},"transaction_costs":{"status":"not_applicable","modeled":"n/a","headline_net_of_costs":null,"detail":null,"quote":null,"source":"front_matter","rule":"R3","candidates":{"front_matter":"n/a","evidence_card":"n/a","evidence_card_quoted":false},"meaning":"the paper reports no trading or execution result"},"results":{"status":"available","evaluation_type":"unclear","findings":[{"finding":"During the Covid crisis, precious metals and pharmaceutical products are identified as causal drivers of the financial network.","kind":"statistical","quote":"Using Kenneth French's business sector return time series, it is revealed that during the Covid crisis, precious metals and pharmaceutical products are causal drivers of the financial network.","support":"full"},{"finding":"The estimated Granger causality network shows high connectivity during crisis periods.","kind":"statistical","quote":"Moreover, the estimated Granger causality network shows a high connectivity during crisis which means that the research presented here can be especially useful to better understand crises in the market by revealing the dominant drivers of the crisis dynamics.","support":"full"},{"finding":"Overall market connectivity mostly stays within the random‑data confidence interval, but certain periods exhibit significant spikes in connectivity.","kind":"statistical","quote":"The results in figure 3 show that the market mostly remains within the range expected from random time series, but some periods exhibit a spike to a significantly high level of network connectivity.","support":"full"},{"finding":"The gradient flow contribution for the 2007 reduced network is γ = 0.98, higher than for any 2020 network.","kind":"theoretical","quote":"This is confirmed by the estimation of the gradient flow contribution for the reduced network which yields γ(2007) = 0.98 and shows an even higher gradient contribution than for any of the 2020 networks.","support":"full"},{"finding":"Top nodes of the hierarchy are estimated with high accuracy even with observation noise up to σobs ≈ 1.","kind":"statistical","quote":"Even after adding observation noise N(0, σobs) to the simulated data, the top nodes of the networks hierarchy are still estimated with a high accuracy beyond the random expectation for noise up to σobs ≈1.","support":"full"}],"reported_performance":null,"out_of_sample":{"method":"none","headline_out_of_sample":null}},"independent_evidence":{"status":"available","supports":[{"paper":{"paper_id":"2601.01783","title":"Dynamic Risk in the U.S. Banking System: An Analysis of Sentiment, Policy Shocks, and Spillover Effects","paper_date":"2026-01-05","authors":["Haibo Wang","Jun Huang","Lutfu S Sua","Jaime Ortiz","Jinshyang Roan","Bahram Alidaee"],"empirical_rigor":7.5,"page_url":"https://thequant.space/flowcharts/dynamic-risk-in-the-u.s.-banking-system-an-analysis-of-sent/"},"question":"Do crisis periods cause a rise in financial network connectivity?","this_paper_finding":"The estimated Granger causality network shows high connectivity during crisis periods.","other_paper_finding":"Total system connectedness surged dramatically during the crisis peak, indicating heightened systemic risk.","other_paper_quote":"Our results provide strong evidence for this contagion channel: total system connectedness surged dramatically during the crisis peak, and we identify SIVB, FRC, and WAL as primary net transmitters of risk while their perceived peers became significant net receivers, a key dynamic indicator of systemic vulnerability that cannot be captured by asset-by-asset analysis.","note":"Both findings report heightened connectivity at the crisis peak – A via Granger‑causality links, B via total system connectedness – same financial market, B focuses on U.S. banks.","confidence":0.601}],"contradicts":[],"qualifies":[],"compared_with":21,"citations":{"cited_by_in_corpus":0,"by_independent_authors":0,"how_cited":{"used_as_baseline":0,"uses_method_or_data":0,"extends":0,"discusses_findings":0,"mentions":0,"unclassified":0},"independent_papers":[],"scope":"Citations between papers in this corpus only; citations from outside it are not counted."}},"implementation":{"status":"available","public":false,"code_url":null,"note":"No repository presented by the authors as their own was found in the paper. Absence here is not proof that no code exists."},"replication":{"status":"none_found","card":null,"note":"No replication of this paper has been run by thequant.space yet."},"extraction":{"text_source":"fulltext","model":"openai/gpt-oss-120b","version":2,"extracted_at":"2026-11-14T19:55:51Z"},"provenance":{"claims":"Everything here is what the authors state in the paper. Nothing was re-run or reproduced by the platform.","quotes":"Assumptions, datasets, cost detail and findings were extracted by a language model. Every quote is verbatim paper text. Where an item has `support`, \"full\" means every figure in it appears in the quoted sentences and \"partial\" means a figure is stated elsewhere in the paper; items with a figure found nowhere in the paper were discarded.","datasets_known":"datasets.known lists well-known vendors and public datasets by a stable id, found by a dictionary over the full text where the sentence says the data is used; it is independent of the model and covers only names in the dictionary.","flags":"transaction_costs.modeled, results.evaluation_type and results.out_of_sample.method are platform classifications (structured-decision models); see https://thequant.space/score-guide/#penalty-flags","comparisons":"supports / contradicts / qualifies are a language model's reading of two papers' quoted findings, restricted to closely related papers that share no author. They are pointers to read, not verified replications, and coverage is partial.","implementation":"code_url is a repository the paper itself presents as its own (rule-scored, ambiguous cases judged by a model), checked for liveness when it was recorded."},"docs":"https://thequant.space/agents/"}