{"paper_id":"2412.11602","title":"Multivariate Distributions in Non-Stationary Complex Systems II: Empirical Results for Correlated Stock Markets","authors":["Anton J. Heckens","Efstratios Manolakis","Cedric Schuhmann","Thomas Guhr"],"paper_date":"2024-12-16","paper_type":"empirical","paper_url":"https://arxiv.org/pdf/2412.11602v2","page_url":"https://thequant.space/flowcharts/multivariate-distributions-in-non-stationary-complex-systems/","scores":{"empirical_rigor":7.5,"math_complexity":8,"quadrant":"Holy Grail"},"proposes":{"status":"available","statement":"The paper applies a Random Matrix Model to multivariate stock return distributions, providing few‑parameter formulae that capture heavy, algebraic tails and non‑stationarity.","quote":"Here, we apply these results to the returns measured in correlated stock markets.","summary":"Multivariate Distributions are needed to capture the correlation structure of complex systems. In previous works, we developed a Random Matrix Model for such correlated multivariate joint probability density functions that accounts for the non-stationarity typically found in complex systems. Here…","strategy":null},"assumptions":{"status":"available","items":[{"assumption":"Returns are calculated from midpoint prices, assuming best ask and bid are available and comparable to transaction prices.","type":"data","stated":"explicit","quote":"From the midpoint price, we calculate the logarithmic returns"},{"assumption":"The set of stocks is continuously traded on every open day and remains in the S&P 500 index throughout 2014.","type":"market","stated":"explicit","quote":"We select stocks being continually traded on every open day while simultaneously being part of the S&P 500 index in 2014."},{"assumption":"Correlation matrices are noisy and are optionally denoised with Ledoit–Wolf shrinkage, but the main analysis assumes the raw empirical matrices.","type":"model","stated":"explicit","quote":"We use Ledoit–Wolf shrinkage [77], which reduces noise in covariance matrices, on the distribution of the aggregated returns"},{"assumption":"Non‑stationarity of correlations is modeled by random matrices, implying that correlation fluctuations drive tail heaviness.","type":"model","stated":"explicit","quote":"The fluctuating correlations in the non‑stationary system are modeled by random matrices."},{"assumption":"Rotating returns into the eigenbasis of the correlation matrix preserves all information and facilitates analysis.","type":"statistical","stated":"explicit","quote":"We rotate the data into the eigenbasis of the correlation matrix."},{"assumption":"Aggregating returns over longer horizons improves statistics without restricting results to the eigenbasis.","type":"statistical","stated":"explicit","quote":"By using the aggregated returns we facilitate the data analysis, but our results are not restricted to the eigenbasis"}]},"datasets":{"status":"available","items":[{"name":"Intraday returns of 479 US stocks","provider":"S&P 500 constituents","kind":"market-prices","access":"public","period":"2014","universe":"479 stocks in the S&P 500 index","quote":"We analyze intraday data of 479 US stocks included in the S&P500 index during the trading year of 2014."},{"name":"Daily adjusted prices of 308 US stocks","provider":"S&P 500 constituents","kind":"market-prices","access":"public","period":"January 23, 1992 to December 31, 2012","universe":"308 stocks listed in the S&P 500 index","quote":"In total, it comprises 308 stocks for a time period ranging from January 23, 1992 to December 31, 2012. This data set lists daily adjusted prices."}],"known":[{"id":"taq","name":"NYSE TAQ","provider":"NYSE","kind":"market-prices","access":"commercial","quote":"Data sets We use the Daily TAQ (Trade and Quote) of the year 2014 from the New York stock exchange (NYSE) [62].","mentions":3,"used_in":1}],"sample":{"start":null,"end":null,"market":null,"asset_class":null,"frequency":null},"data_resolution":"tick"},"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":"pricing-calibration","findings":[{"finding":"Non‑stationary fluctuations of correlations make the return tails heavier, especially for longer time intervals.","kind":"statistical","quote":"The non‑stationary fluctuations of the correlations lift the tails."},{"finding":"Four multivariate model distributions with heavy, mostly algebraic tails fit the empirical data well.","kind":"theoretical","quote":"We obtained four multivariate model distributions with heavy, mostly algebraic tails which we fit to the data."},{"finding":"Aggregated‑return fits using Algebraic‑Algebraic, Algebraic‑Gaussian and Gaussian‑Algebraic models outperform the Gaussian‑Gaussian case.","kind":"performance","quote":"the fits for ⟨p⟩(aggr)GA(er), ⟨p⟩(aggr)AG(er) and ⟨p⟩(aggr)AA(er) outperform the one"},{"finding":"Fit parameters increase with larger return horizons, consistent with earlier studies.","kind":"statistical","quote":"The fit parameters tend to be larger for a larger return horizon ∆t"},{"finding":"Rotation into the eigenbasis captures full correlation information and does not limit applicability to that basis.","kind":"theoretical","quote":"As opposed to the univariate distributions of the original, unrotated returns, the univariate distributions of the rotated returns depend on the eigenvalues of the correlation matrix and thus contain the full information on the correlated system."}],"reported_performance":null,"out_of_sample":{"method":"none","headline_out_of_sample":null}},"independent_evidence":{"status":"available","supports":[{"paper":{"paper_id":"2302.08208","title":"A Look at Financial Dependencies by Means of Econophysics and Financial Economics","paper_date":"2023-02-16","authors":["M. Raddant","T. Di Matteo"],"empirical_rigor":5.5,"page_url":"https://thequant.space/flowcharts/a-look-at-financial-dependencies-by-means-of-econophysics-and-financial/"},"question":"Do non‑stationary correlation fluctuations lead to heavier return tails that inflate the significance of correlation tests?","this_paper_finding":"Non‑stationary fluctuations of correlations make the return tails heavier, especially for longer time intervals.","other_paper_finding":"Parametric tests tend to overstate the significance of correlation coefficients because the tails of the permutated return distribution contain more extremes than those of independent normal noise.","other_paper_quote":"When we compare these tails with the corresponding distributions of inde-pendent normally distributed noise, we observe slightly too many extremes in the correlation coeﬃcients of the permutated returns, especially for small ∆t. Hence, relying on a parametric test for the correlation coeﬃcients can in this case lead to an overstatement of their signiﬁcance (see also Kowalski, 1972).","note":"A finds parametric tests overstate significance due to extreme tails in permuted returns; B shows non‑stationary correlation fluctuations make tails heavier, especially over longer intervals, supporting A.","confidence":0.561},{"paper":{"paper_id":"2404.11722","title":"Beyond the Bid-Ask: Strategic Insights into Spread Prediction and the Global Mid-Price Phenomenon","paper_date":"2024-04-17","authors":["Yifan He","Abootaleb Shirvani","Barret Shao","Svetlozar Rachev","Frank Fabozzi"],"empirical_rigor":7,"page_url":"https://thequant.space/flowcharts/beyond-the-bid-ask-strategic-insights-into-spread-predictio/"},"question":"Do the log‑return distributions of TMOBBAS and GMP exhibit heavy‑tailed behavior?","this_paper_finding":"Four multivariate model distributions with heavy, mostly algebraic tails fit the empirical data well.","other_paper_finding":"Log‑return distributions of both TMOBBAS and GMP exhibit heavy tails and significant deviations from normality.","other_paper_quote":"CIs for each depth. This revealed that the log‑return distributions for both TMOBBAS and GMP have heavy tails, indicative of significant deviations from normality.","note":"Both findings report heavy tails: A notes heavy‑tailed log‑returns for TMOBBAS and GMP; B finds four multivariate models with algebraic (heavy) tails fit the data, supporting A.","confidence":0.532}],"contradicts":[],"qualifies":[],"compared_with":42,"citations":{"cited_by_in_corpus":1,"by_independent_authors":0,"how_cited":{"used_as_baseline":0,"uses_method_or_data":0,"extends":0,"discusses_findings":0,"mentions":0,"unclassified":1},"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":1,"extracted_at":"2026-11-14T17:49:31Z"},"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/"}