{"paper_id":"2511.08616","title":"Reasoning on Time-Series for Financial Technical Analysis","authors":["Kelvin J. L. Koa","Jan Chen","Yunshan Ma","Huanhuan Zheng","Tat-Seng Chua"],"paper_date":"2025-11-06","paper_type":"empirical","paper_url":"https://arxiv.org/pdf/2511.08616v1","page_url":"https://thequant.space/flowcharts/reasoning-on-time-series-for-financial-technical-analysis/","scores":{"empirical_rigor":6,"math_complexity":6.5,"quadrant":"Holy Grail"},"proposes":{"status":"available","statement":"The paper proposes the Verbal Technical Analysis (VTA) framework that combines a backbone time-series model with a reasoning LLM to produce interpretable stock forecasts.","quote":"we propose our Verbal Technical Analysis (VTA) framework, which combines a backbone time-series model (which we termed as “latent thinking”) with a reasoning LLM (termed as “verbal reasoning”) to produce interpretable stock time-series forecasts.","summary":"While Large Language Models have been used to produce interpretable stock forecasts, they mainly focus on analyzing textual reports but not historical price data, also known as Technical Analysis. This task is challenging as it switches between domains: the stock price inputs and outputs lie in the…","strategy":"Verbal Technical Analysis (VTA)"},"assumptions":{"status":"available","items":[{"assumption":"The conditional probability puncond is set to 0.3 and the guidance scale s to 0.1 for the model.","type":"model","stated":"explicit","quote":"For the forecasting model, we use GPT-2 (Radford et al., 2019) as the base model. For hyperparameters, we set the conditional probability puncond to 0.3 and the guidance scale s to 0.1."},{"assumption":"Stocks are selected bias‑free from well‑known indices (Dow Jones, FTSE China A50, EURO STOXX 50).","type":"data","stated":"explicit","quote":"To ensure a bias-free selection, we choose the stocks from well-known indices, i.e., the Dow Jones, the FTSE China A50 Index and the EURO STOXX 50."},{"assumption":"The forecasting loss is mean‑squared error between predicted and ground‑truth series.","type":"model","stated":"explicit","quote":"The model is then fine-tuned using MSE loss with the ground truth series y."},{"assumption":"Portfolio construction uses Markowitz optimization over the 10‑day predictions.","type":"execution","stated":"explicit","quote":"This is achieved by performing Markowitz optimization (Markowitz, 1952) across the 10-day predictions."}]},"datasets":{"status":"available","items":[{"name":"ACL18 StockNet dataset","provider":"Xu & Cohen, 2018","kind":"benchmark","access":"public","period":"01/09/2012 to 01/09/2017","universe":"88 U.S. stocks representing top 8‑10 companies by market cap in each major industry","quote":"The first is the ACL18 StockNet dataset (Xu & Cohen, 2018), which includes historical price data for 88 U.S. stocks that are selected to represent the top 8-10 companies by market capitalization in each of the major industries. The dataset spans the period of 01/09/2012 to 01/09/2017."},{"name":"U.S., Chinese, and European market data","provider":null,"kind":"market-prices","access":"unknown","period":"01/01/2024 to 01/01/2025","universe":"Stocks from Dow Jones, FTSE China A50 Index, EURO STOXX 50","quote":"To further show the generalization ability of the model, we also collect additional stock data from across the U.S., Chinese, and European markets for testing."}],"known":[],"sample":{"start":null,"end":null,"market":"global equity markets (U.S., Chinese, European)","asset_class":"equities","frequency":"daily"},"data_resolution":"daily"},"transaction_costs":{"status":"available","modeled":"no","headline_net_of_costs":false,"detail":null,"quote":null,"source":"front_matter","rule":"R1","candidates":{"front_matter":"no","evidence_card":"unclear","evidence_card_quoted":false},"meaning":"a trading or execution result is reported without costs"},"results":{"status":"available","evaluation_type":"trading-pnl","findings":[{"finding":"VTA achieves the best performance among all models in both MSE and MAE on the test sets.","kind":"performance","quote":"Our proposed VTA model demonstrates the best performance in stock forecasting in both MSE and MAE."},{"finding":"VTA obtains the highest Sharpe ratio among all evaluated models in portfolio optimization.","kind":"performance","quote":"VTA achieves the highest Sharpe ratio among all models, which represents the risk-adjusted returns."},{"finding":"The second RL fine‑tuning stage improves average MSE by about 20.3% over the base model.","kind":"statistical","quote":"the second RL fine-tuning, which uses the same Time-GRPO objective, produces an average improvement of 20.3%."}],"reported_performance":{"strategy":"Verbal Technical Analysis (VTA)","sharpe":1.719,"sharpe_benchmark":null,"annual_return_pct":null,"total_return_pct":null,"max_drawdown_pct":8.83},"out_of_sample":{"method":"holdout","headline_out_of_sample":true}},"independent_evidence":{"status":"none_found","supports":[],"contradicts":[],"qualifies":[],"compared_with":31,"citations":{"cited_by_in_corpus":1,"by_independent_authors":1,"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":true,"code_url":"https://github.com/chen-jan/VTA","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":"2026-11-14T17:22:38Z"},"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/"}