{"paper_id":"2608.12283","title":"Large Language Model-Driven Small-Capitalization Trading: Integrating Financial News Sentiment, Macroeconomic Indicators, and Technical Signals","authors":["Alireza Kargarzadeh","Nariman Khaledian","Navid Parvini","Arman Khaledian"],"paper_date":"2026-08-12","paper_type":"empirical","paper_url":"https://arxiv.org/pdf/2608.12283","page_url":"https://thequant.space/flowcharts/large-language-model-driven-small-capitalization-trading-integrating-financial/","scores":{"empirical_rigor":8,"math_complexity":6,"quadrant":"Holy Grail"},"proposes":{"status":"available","statement":"The paper proposes an uncertainty‑aware portfolio construction that feeds model‑predicted risk, split into aleatoric and epistemic components, directly into the covariance matrix used by allocators.","quote":"We study an uncertainty-aware construction that feeds model-predicted risk -- decomposed into aleatoric and epistemic components -- directly into the covariance matrix of portfolio allocators, rather than treating portfolio risk as fixed or adjusting only expected returns.","summary":"Large language models can extract richer signals from financial news than fixed sentiment lexicons, and recent work has explored feeding such signals into portfolio construction. We study an uncertainty-aware construction that feeds model-predicted risk – decomposed into aleatoric and epistemic…","strategy":"Pure beta with GPT-4o mini sentiment, Student-t target, 40-day holding, risk par"},"assumptions":{"status":"available","items":[{"assumption":"The aleatoric risk is modeled with either a Gaussian or Student‑t distribution and the model is trained by maximizing likelihood under that distribution.","type":"model","stated":"explicit","quote":"The model is trained by maximizing the likelihood of realized returns under its own predicted aleatoric distribution, Gaussian or Student-t (full training objective in Appendix E), so risk is learned jointly with return rather than fit afterward from residuals."},{"assumption":"All data used for universe selection, standardization, model fitting, and early stopping are confined to an in‑sample period ending 31 Dec 2024.","type":"data","stated":"explicit","quote":"All universe selection, standard‑ization, model fitting, and early stopping use only the in‑sample period ending December 31, 2024."},{"assumption":"Transaction costs are represented by a set of fixed basis‑point scenarios plus a constant 5 bps one‑way slippage.","type":"execution","stated":"explicit","quote":"Transaction-cost scenarios are {0, 1, 2, 5, 10, 20, 50, 100} basis points, with a fixed 5 bps one-way slippage component in addition to the reported transaction-cost setting."},{"assumption":"Holding periods are limited to the discrete set {1,2,3,5,10,20,40,60} trading days.","type":"other","stated":"explicit","quote":"Holding periods are H ∈{1, 2, 3, 5, 10, 20, 40, 60} trading days."},{"assumption":"The predictive covariance matrix is constructed by summing aleatoric and epistemic covariances, with epistemic covariance zero when S=1.","type":"model","stated":"explicit","quote":"We combine both into a single predictive covariance, ˆΣt = ˆΣA t + ˆΣE t (6) extending it from a single-signal quantity to a full N × N cross-asset covariance matrix. With S = 1, epistemic covariance is exactly zero, recovering a model with no explicit account of its own uncertainty."},{"assumption":"Backtesting is performed on a single out‑of‑sample calendar year (2025), so annualized metrics are summaries of one test year.","type":"data","stated":"explicit","quote":"Because the out-of-sample period is one calendar year, annualized metrics should be read as annualized summaries of a single test year rather than as estimates over many independent years."}]},"datasets":{"status":"available","items":[{"name":"Russell 2000 equities daily OHLCV","provider":null,"kind":"market-prices","access":"unknown","period":"Oct 2 2023 – Dec 31 2025","universe":"Russell 2000 constituents","quote":"We evaluate the pipeline on Russell 2000 equities with daily OHLCV data from October 2, 2023 through December 31, 2025"},{"name":"Scored news articles","provider":null,"kind":"text-news","access":"unknown","period":"Oct 1 2023 – Dec 31 2025","universe":null,"quote":"scored news from October 1, 2023 through December 31, 2025"},{"name":"Macro‑indicator panel","provider":null,"kind":"macro","access":"unknown","period":"Jan 1 2022 – Jan 1 2026","universe":null,"quote":"Macro-beta selection additionally uses a macro‑indicator panel spanning January 1, 2022 through January 1, 2026"}],"known":[],"sample":{"start":"2025","end":"2026-01","market":"Russell 2000","asset_class":"equities","frequency":"daily"},"data_resolution":"daily"},"transaction_costs":{"status":"available","modeled":"yes","headline_net_of_costs":true,"detail":"Costs are modeled as explicit basis‑point scenarios (0–100 bps) plus a fixed 5 bps one‑way slippage.","quote":"Transaction-cost scenarios are {0, 1, 2, 5, 10, 20, 50, 100} basis points, with a fixed 5 bps one-way slippage component in addition to the reported transaction-cost setting.","source":"front_matter","rule":"R3","candidates":{"front_matter":"yes","evidence_card":"yes","evidence_card_quoted":true},"card_detail":"Costs are modeled as explicit basis‑point scenarios (0–100 bps) plus a fixed 5 bps one‑way slippage.","card_quote":"Transaction-cost scenarios are {0, 1, 2, 5, 10, 20, 50, 100} basis points, with a fixed 5 bps one-way slippage component in addition to the reported transaction-cost setting.","meaning":"costs are part of the reported headline results"},"results":{"status":"available","evaluation_type":"trading-pnl","findings":[{"finding":"The pure‑beta leg with GPT‑4o mini sentiment, Student‑t target, 40‑day holding, and risk‑parity allocation achieves a Sharpe of 2.33 at 100 bps transaction costs.","kind":"performance","quote":"The strongest conservative row is pure beta with GPT-4o mini sentiment, a Student-t target, a 40-day holding period, and risk parity allocation, reaching Sharpe 2.33 at 100 bps."},{"finding":"Separated pure‑alpha and pure‑beta legs usually dominate the beta intersection on Sharpe and return across the holding‑period grid.","kind":"performance","quote":"Across the full holding-period grid, the separated pure-alpha and pure-beta legs usually dominate the beta intersection on Sharpe and return."},{"finding":"Pure‑alpha leads at 5, 10, and 60 trading days; pure‑beta leads at 20 and 40 days, with pure‑beta’s one‑day edge disappearing under 100 bps costs.","kind":"performance","quote":"Pure alpha leads at 5, 10, and 60 trading days; pure beta leads at 20 and 40 days. The one-day result is cost-sensitive: pure beta is strongest at low costs, but the edge disappears under the 100 bps stress case."},{"finding":"Transaction‑cost robustness shows pure‑alpha remains strongest from zero to 100 bps at a 60‑day horizon, with its lead widening as costs rise.","kind":"performance","quote":"pure alpha remains the strongest filtered regime from zero costs through 100 bps; if anything, its lead over pure beta and the beta intersection widens rather than narrows as costs rise."},{"finding":"The way the tradable set is defined (pure‑alpha, pure‑beta, or their intersection) matters at least as much as the sentiment model or allocator choice.","kind":"other","quote":"The central result is that the way the tradable set is defined matters at least as much as the sentiment backend or allocator."}],"reported_performance":{"strategy":"Pure beta with GPT-4o mini sentiment, Student-t target, 40-day holding, risk par","sharpe":2.33,"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":29,"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":"2026-11-14T17:05:40Z"},"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/"}