{"paper_id":"2509.11970","title":"Sentiment Feedback in Equity Markets: Asymmetries, Retail Heterogeneity, and Structural Calibration","authors":["Lucas Marques Sneller"],"paper_date":"2025-09-15","paper_type":"empirical","paper_url":"https://arxiv.org/pdf/2509.11970v1","page_url":"https://thequant.space/flowcharts/sentiment-feedback-in-equity-markets-asymmetries-retail-he/","scores":{"empirical_rigor":7.2,"math_complexity":6.5,"quadrant":"Holy Grail"},"proposes":{"status":"available","statement":"The paper proposes a two‑parameter (κ, ρ) structural feedback model to capture how sentiment shocks amplify equity returns and calibrates it to observed impulse responses.","quote":"In summary, our contributions lie in empirically demonstrating three interrelated phenomena","summary":"We study how sentiment shocks propagate through equity returns and investor clientele using four independent proxies with sign-aligned kappa-rho parameters. A structural calibration links a one standard deviation innovation in sentiment to a pricing impact of 1.06 basis points with persistence…","strategy":"D10‑D1 long‑short portfolio"},"assumptions":{"status":"available","items":[{"assumption":"Sentiment shocks are AR(1) innovations standardized to unit variance.","type":"model","stated":"explicit","quote":"Monthly sentiment shocks are AR(1) innovations from the University of Michigan Consumer Sentiment Index (UMCSENT), standardized to unit variance."},{"assumption":"Retail intensity is proxied by the fraction of retail trading volume estimated with the Lee–Ready algorithm.","type":"data","stated":"explicit","quote":"Primary proxy is the fraction of retail trading volume estimated from TAQ data using the Lee–Ready algorithm."},{"assumption":"Breadth is measured as the fraction of reporting managers holding a stock, expanded to monthly frequency.","type":"data","stated":"explicit","quote":"We construct breadth from LSEG S34 Type 3 quarterly holdings as the fraction of reporting managers holding stock i at quarter t."},{"assumption":"Optionability is a binary flag based on first listing date from CBOE historical data.","type":"data","stated":"explicit","quote":"Optionable flag based on first listing date from CBOE historical data."},{"assumption":"Transaction costs are applied as flat one‑way fees of 0, 5, or 10 basis points per trade.","type":"execution","stated":"explicit","quote":"... Net Sharpe ratios account for transaction costs."}]},"datasets":{"status":"available","items":[{"name":"UMCSENT sentiment shocks","provider":"University of Michigan","kind":"macro","access":"public","period":"1990–2024","universe":null,"quote":"Monthly sentiment shocks are AR(1) innovations from the University of Michigan Consumer Sentiment Index (UMCSENT), standardized to unit variance. Sample period: 1990–2024 (420 monthly observations)."},{"name":"LSEG S34 breadth data","provider":"LSEG","kind":"alternative","access":"commercial","period":"1990–2024","universe":"~7,725 firms per month","quote":"We construct breadth from LSEG S34 Type 3 quarterly holdings as the fraction of reporting managers holding stock i at quarter t.","known_id":"refinitiv"},{"name":"Retail intensity proxy","provider":"TAQ (via Lee–Ready algorithm)","kind":"alternative","access":"commercial","period":"2003–2024","universe":"~2,000 firms per month","quote":"Primary proxy is the fraction of retail trading volume estimated from TAQ data using the Lee–Ready algorithm. Sample: 2003–2024, ∼2,000 firms per month.","known_id":"taq"},{"name":"Optionability flag","provider":"CBOE historical data","kind":"options","access":"commercial","period":"through 2024","universe":"≈60% of sample firms are optionable by 2024","quote":"Optionable flag based on first listing date from CBOE historical data. Non‑optionable stocks defined as those without options trading history. Coverage: ∼60% of sample firms are optionable by 2024.","known_id":"cboe"},{"name":"Market‑wide VIX and asset‑specific VX series","provider":"CBOE (VIX) and other providers","kind":"macro","access":"public","period":"1990–2024","universe":null,"quote":"Implied volatility series. Market‑wide VIX (near full coverage), asset‑specific VX series where available (∼40% coverage), OVX (oil) and VXEEM (emerging markets) for robustness.","known_id":"cboe"}],"known":[{"id":"refinitiv","name":"Refinitiv / LSEG (Datastream, Eikon, Tick History)","provider":"LSEG","kind":"market-prices","access":"commercial","quote":"We use licensed data from LSEG (S34/13F), CRSP/Compustat, CBOE, and TAQ (for sub-samples with retail proxies).","mentions":3,"used_in":3},{"id":"crsp","name":"CRSP","provider":"Center for Research in Security Prices","kind":"market-prices","access":"commercial","quote":"We use licensed data from LSEG (S34/13F), CRSP/Compustat, CBOE, and TAQ (for sub-samples with retail proxies).","mentions":5,"used_in":1},{"id":"cboe","name":"Cboe","provider":"Cboe Global Markets","kind":"options","access":"commercial","quote":"We use licensed data from LSEG (S34/13F), CRSP/Compustat, CBOE, and TAQ (for sub-samples with retail proxies).","mentions":3,"used_in":1},{"id":"taq","name":"NYSE TAQ","provider":"NYSE","kind":"market-prices","access":"commercial","quote":"We use licensed data from LSEG (S34/13F), CRSP/Compustat, CBOE, and TAQ (for sub-samples with retail proxies).","mentions":3,"used_in":1},{"id":"compustat","name":"Compustat","provider":"S&P Global","kind":"fundamentals","access":"commercial","quote":"We use licensed data from LSEG (S34/13F), CRSP/Compustat, CBOE, and TAQ (for sub-samples with retail proxies).","mentions":1,"used_in":1}],"sample":{"start":"1990","end":"2024","market":"U.S. equities","asset_class":"equities","frequency":"monthly"},"data_resolution":"tick"},"transaction_costs":{"status":"available","modeled":"yes","headline_net_of_costs":true,"detail":"Flat one‑way fees of 0, 5, or 10 bps per trade are applied; net Sharpe ratios are reported after subtracting these costs.","quote":"... Net Sharpe ratios account for transaction costs.","source":"front_matter","rule":"R3","candidates":{"front_matter":"yes","evidence_card":"yes","evidence_card_quoted":true},"card_detail":"Flat one‑way fees of 0, 5, or 10 bps per trade are applied; net Sharpe ratios are reported after subtracting these costs.","card_quote":"... Net Sharpe ratios account for transaction costs.","meaning":"costs are part of the reported headline results"},"results":{"status":"available","evaluation_type":"trading-pnl","findings":[{"finding":"A one‑standard‑deviation sentiment shock produces a pricing impact of 1.06 bps with persistence ρ = 0.940, implying an 11.2‑month half‑life.","kind":"pricing-calibration","quote":"A one‑standard deviation innovation in sentiment to a pricing impact of 1.06 basis points with persistence parameter rho = 0.940, yielding a half-life of 11.2 months."},{"finding":"The D10‑D1 long‑short portfolio earns 4.0 bps per month (up to 13 bps in some specifications) with Sharpe ratios ranging from 0.18 to 0.85 after realistic transaction costs.","kind":"performance","quote":"Cross‑sectionally, a simple D10‑D1 portfolio earns 4.0 basis points per month with Sharpe ratios of 0.18‑0.85, consistent with tradable exposure to the sentiment factor."},{"finding":"Positive sentiment shocks have larger and more persistent return responses than negative shocks (amplification asymmetry).","kind":"statistical","quote":"positive sentiment innovations transmit more strongly than negative shocks (amplification asymmetry)."},{"finding":"In high‑VIX months the impact of a sentiment shock is larger but decays faster; in low‑VIX months the impact is smaller but more persistent.","kind":"statistical","quote":"a one‑standard‑deviation sentiment innovation has a larger impact on prices on impact during high‑VIX months but the effect decays relatively quickly, whereas in low‑VIX months the initial impact is smaller yet the persistence (half‑life of the sentiment impact) is markedly longer."},{"finding":"Triple interaction of low breadth, high VIX, and retail intensity yields amplification of 12‑31 bps, demonstrating economically meaningful effects in constrained market segments.","kind":"statistical","quote":"The triple interaction with low breadth and high VIX reaches 12‑31 bps, demonstrating economically meaningful amplification in constrained market segments."}],"reported_performance":{"strategy":"D10‑D1 long‑short portfolio","sharpe":0.85,"sharpe_benchmark":null,"annual_return_pct":null,"total_return_pct":null,"max_drawdown_pct":null},"out_of_sample":{"method":"holdout","headline_out_of_sample":true}},"independent_evidence":{"status":"none_found","supports":[],"contradicts":[],"qualifies":[],"compared_with":32,"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":1,"extracted_at":"2026-11-14T17:27:14Z"},"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/"}