{"paper_id":"2407.20380","title":"Inferring financial stock returns correlation from complex network analysis","authors":["Ixandra Achitouv"],"paper_date":"2024-07-29","paper_type":"empirical","paper_url":"https://arxiv.org/pdf/2407.20380v1","page_url":"https://thequant.space/flowcharts/inferring-financial-stock-returns-correlation-from-complex-n/","scores":{"empirical_rigor":6,"math_complexity":7.5,"quadrant":"Holy Grail"},"proposes":{"status":"available","statement":"The paper proposes combining random matrix theory with complex network analysis to model stock return correlations via correlated GBM and use the simulated market walks to build a better performing portfolio.","quote":"In this work we combine the RMT analysis with the network analysis of the S&P500 stocks to better understand the collective modes present in the stock returns correlations.","summary":"Financial stock returns correlations have been studied in the prism of random matrix theory, to distinguish the signal from the “noise”. Eigenvalues of the matrix that are above the rescaled Marchenko Pastur distribution can be interpreted as collective modes behavior while the modes under are…","strategy":null},"assumptions":{"status":"available","items":[{"assumption":"Stock prices follow a log‑normal distribution with constant volatility and drift in the GBM model.","type":"model","stated":"explicit","quote":"The GBM model assumes that stock prices are log-normally distributed."},{"assumption":"Correlation parameters (c_eff) are derived from network properties such as clustering coefficient, eigenvector centrality and Louvain communities.","type":"model","stated":"explicit","quote":"The correlations are modeled by hyper-parameters that are computed from the network properties."},{"assumption":"Only S&P500 constituents present for the entire sample period are used, resulting in 485 stocks.","type":"data","stated":"explicit","quote":"We clear out stocks that were not present in the entire time range, ending up with 485 stocks and 1258 days of closing values for each stock."},{"assumption":"Historical parameters S0, µ and σ are estimated from the observed price series for each stock.","type":"data","stated":"explicit","quote":"The parameters S0, µ and σ are computed from the historical data for every stocks."},{"assumption":"Portfolio rebalancing is performed at fixed time scales dT without accounting for transaction costs.","type":"execution","stated":"implied","quote":"period ... we re‑compute the weights on different time scales dT=[252; 252/2; 252/3; 252/4; 252/6; 252/8]"}]},"datasets":{"status":"available","items":[{"name":"S&P500 daily closing prices","provider":"Yahoo Finance","kind":"market-prices","access":"public","period":"2019-01-01 to 2024-01-01","universe":"S&P500 constituents (485 after cleaning)","quote":"In what follows we consider the stocks from the S&P500 from 01-01-2019 to 01-01-2024 downloaded from Yahoo finance.","known_id":"yahoo-finance"}],"known":[{"id":"yahoo-finance","name":"Yahoo Finance","provider":"Yahoo","kind":"market-prices","access":"public","quote":"Datasets In what follows we consider the stocks from the S&P500 from 01-01-2019 to 01-01-2024 downloaded from Yahoo finance.","mentions":2,"used_in":2}],"sample":{"start":null,"end":null,"market":null,"asset_class":null,"frequency":null},"data_resolution":"daily"},"transaction_costs":{"status":"available","modeled":"no","headline_net_of_costs":null,"detail":null,"quote":null,"source":"front_matter","rule":"R1","candidates":{"front_matter":"no","evidence_card":"no","evidence_card_quoted":false},"meaning":"a trading or execution result is reported without costs"},"results":{"status":"available","evaluation_type":"trading-pnl","findings":[{"finding":"The simulated market GBM portfolio yields a cumulative return of 39% versus 26% for the baseline, a 50% improvement.","kind":"performance","quote":"The cumulative return for the baseline scenario is 26% while for the simulation it is 39%, a 50% improvement."},{"finding":"For rebalancing intervals dT ≤ 84 days the simulated walks outperform the baseline, giving a 6% higher return (69% vs 65%).","kind":"performance","quote":"For this time scale the return of the baseline is 65% and for the simulated walks it is 69%, a 6% improvement."},{"finding":"The returns correlation matrix is dominated by stocks with high eigenvector centrality and clustering in the network.","kind":"statistical","quote":"We find that the returns correlation matrix is dominated by stocks with high eigenvector centrality and clustering found in the network."}],"reported_performance":null,"out_of_sample":{"method":"walk-forward","headline_out_of_sample":null}},"independent_evidence":{"status":"available","supports":[{"paper":{"paper_id":"2301.04455","title":"Utilizing Technical Data to Discover Similar Companies in Dhaka Stock Exchange","paper_date":"2023-01-11","authors":["Tashreef Muhammad","Tahsin Aziz","Mohammad Shafiul Alam"],"empirical_rigor":4.5,"page_url":"https://thequant.space/flowcharts/utilizing-technical-data-to-discover-similar-companies-in-dhaka-stock-exchange/"},"question":"Which stocks form clusters with high centrality in return correlation networks?","this_paper_finding":"The returns correlation matrix is dominated by stocks with high eigenvector centrality and clustering in the network.","other_paper_finding":"A correlation matrix based on closing‑price returns reveals that companies cluster together, forming visible cliques in the network.","other_paper_quote":"The stronger the correlation, the stronger is the connection between them. It can be seen easily from Figure 2 that not all the companies in DSE have similar connections.","note":"Both findings use return correlation matrices to reveal clustered groups; B adds that stocks with high eigenvector centrality dominate the matrix, supporting A’s observed cliques.","confidence":0.67}],"contradicts":[],"qualifies":[],"compared_with":27,"citations":{"cited_by_in_corpus":2,"by_independent_authors":0,"how_cited":{"used_as_baseline":0,"uses_method_or_data":0,"extends":0,"discusses_findings":0,"mentions":0,"unclassified":2},"independent_papers":[],"scope":"Citations between papers in this corpus only; citations from outside it are not counted."}},"implementation":{"status":"available","public":true,"code_url":"https://github.com/IxandraAchitouv","note":null},"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":1,"extracted_at":"2026-11-14T18:05:09Z"},"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/"}