{"paper_id":"2503.04164","title":"CoFinDiff: Controllable Financial Diffusion Model for Time Series Generation","authors":["Yuki Tanaka","Ryuji Hashimoto","Takehiro Takayanagi","Zhe Piao","Yuri Murayama","Kiyoshi Izumi"],"paper_date":"2025-03-06","paper_type":"empirical","paper_url":"https://arxiv.org/pdf/2503.04164v1","page_url":"https://thequant.space/flowcharts/cofindiff-controllable-financial-diffusion-model-for-time-s/","scores":{"empirical_rigor":8.5,"math_complexity":7,"quadrant":"Holy Grail"},"proposes":{"status":"available","statement":"The paper proposes CoFinDiff, a conditional diffusion model for generating synthetic financial time series conditioned on trends and realized volatility.","quote":"We propose CoFinDiff (Controllable Financial Diffusion model), a synthetic financial data generation model based on conditional diffusion models that accept conditions about the synthetic time series.","summary":"The generation of synthetic financial data is a critical technology in the financial domain, addressing challenges posed by limited data availability. Traditionally, statistical models have been employed to generate synthetic data. However, these models fail to capture the stylized facts commonly…","strategy":null},"assumptions":{"status":"available","items":[{"assumption":"Missing price data points are filled using the price from the previous time step.","type":"data","stated":"explicit","quote":"Missing data points were added using the price from the previous time step."},{"assumption":"Extreme events are up‑sampled by replicating the top 25%, 10%, and 5% of absolute trend values five times each.","type":"model","stated":"explicit","quote":"Data corresponding to the top 25%, 10%, and 5% of absolute trend values were replicated five times each."},{"assumption":"The conditional diffusion model uses cross‑attention to incorporate price‑derived conditions.","type":"model","stated":"explicit","quote":"By incorporating conditions derived from price data into the conditional diffusion model via cross‑attention, CoFinDiff learns the relationships between the conditions and the data."},{"assumption":"Trading hours are 5 hours per day for the 1‑minute data.","type":"market","stated":"explicit","quote":"The trading hours per day was 5 hours, with data spanning from January 1, 2015, to December 31, 2021."}]},"datasets":{"status":"available","items":[{"name":"1‑minute FLEX full historical data","provider":"Japan Exchange Group, Inc.","kind":"market-prices","access":"commercial","period":"2015-01-01 to 2021-12-31","universe":"Asahi Kasei, Mitsubishi Chemical, Daiichi Sankyo, Eneos, Toshiba, Sony (six stocks)","quote":"This study used 1‑minute FLEX full historical data provided by Japan Exchange Group, Inc. The trading hours per day was 5 hours, with data spanning from January 1, 2015, to December 31, 2021."}],"known":[],"sample":{"start":null,"end":null,"market":null,"asset_class":null,"frequency":null},"data_resolution":"intraday-bars"},"transaction_costs":{"status":"not_extracted","modeled":"unclear","headline_net_of_costs":null,"detail":null,"quote":null,"source":"none","rule":"R0","candidates":{"front_matter":"unclear","evidence_card":"unclear","evidence_card_quoted":false},"meaning":"the classifier was not confident"},"results":{"status":"available","evaluation_type":"pricing-calibration","findings":[{"finding":"Synthetic data generated by CoFinDiff capture stylized facts such as fat tails and volatility clustering.","kind":"statistical","quote":"the property of stylized facts, traditionally observed in financial time series data."},{"finding":"CoFinDiff accurately meets specified trend and realized volatility conditions, outperforming vanilla GAN on volatility accuracy.","kind":"performance","quote":"Regarding trend conditions, both CoFinDiff and the Wasserstein GAN satisfy the conditions with similar high accuracy. Contrarily, for realized volatility conditions, CoFinDiff generates synthetic data with higher accuracy than the Wasserstein GAN."},{"finding":"CoFinDiff achieves the highest diversity among compared models, especially under extreme‑event conditions.","kind":"statistical","quote":"CoFinDiff achieves the highest diversity in synthetic data."},{"finding":"Models trained on CoFinDiff‑generated data improve performance on deep hedging tasks under upward‑trend and high‑volatility scenarios.","kind":"performance","quote":"under case (i) that assumes rising underlying asset prices and case (ii) that assumes high volatility of the underlying asset, the model trained on CoFinDiff synthetic data yielded the highest performance."},{"finding":"When evaluated on the full test dataset, models trained on real data still outperform synthetic‑data‑trained models.","kind":"performance","quote":"When all available data served as test data, the model trained on real data achieved the best performance."}],"reported_performance":null,"out_of_sample":{"method":"holdout","headline_out_of_sample":null}},"independent_evidence":{"status":"none_found","supports":[],"contradicts":[],"qualifies":[],"compared_with":58,"citations":{"cited_by_in_corpus":3,"by_independent_authors":2,"how_cited":{"used_as_baseline":0,"uses_method_or_data":0,"extends":0,"discusses_findings":0,"mentions":2,"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":"2026-11-14T17:43:13Z"},"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/"}