{"paper_id":"2512.21791","title":"Synthetic Financial Data Generation for Enhanced Financial Modelling","authors":["Christophe D. Hounwanou","Yae Ulrich Gaba","Pierre Ntakirutimana"],"paper_date":"2025-12-25","paper_type":"empirical","paper_url":"https://arxiv.org/pdf/2512.21791v1","page_url":"https://thequant.space/flowcharts/synthetic-financial-data-generation-for-enhanced-financial-m/","scores":{"empirical_rigor":9,"math_complexity":8.5,"quadrant":"Holy Grail"},"proposes":{"status":"available","statement":"The paper proposes a unified multi-criteria evaluation framework for synthetic financial data covering distributional similarity, temporal coherence, volatility patterns, downstream task performance, and visual structure.","quote":"A multi-criteria evaluation framework assessing distributional similarity, temporal coherence, volatility patterns, downstream task performance, and visual structure via PCA.","summary":"Data scarcity and confidentiality in finance often impede model development and robust testing. This paper presents a unified multi-criteria evaluation framework for synthetic financial data and applies it to three representative generative paradigms: the statistical ARIMA-GARCH baseline…","strategy":null},"assumptions":{"status":"available","items":[{"assumption":"Using a single asset provides a controlled setting for comparison.","type":"data","stated":"explicit","quote":"Using a single asset provides a controlled setting for comparison."},{"assumption":"Log‑returns transformation yields a stationary series suitable for models.","type":"model","stated":"implied","quote":"The raw price series was converted into a stationary and numerically stable representation suitable for both econometric and deep learning models."},{"assumption":"Temporal train/validation/test split avoids forward‑looking leakage.","type":"execution","stated":"explicit","quote":"We adopt a temporal split to avoid forward-looking leakage."},{"assumption":"Synthetic datasets are generated with the same size as the real dataset.","type":"data","stated":"implied","quote":"Each model generates synthetic datasets matched in size to the real dataset."}]},"datasets":{"status":"available","items":[{"name":"S&P 500 daily closing prices","provider":"S&P 500 index","kind":"market-prices","access":"public","period":"January 2000 to March 2024","universe":"S&P 500 index","quote":"We use daily closing prices of the S&P 500 index from January 2000 to March 2024."}],"known":[],"sample":{"start":null,"end":null,"market":null,"asset_class":null,"frequency":null},"data_resolution":"daily"},"transaction_costs":{"status":"not_applicable","modeled":"n/a","headline_net_of_costs":null,"detail":null,"quote":null,"source":"front_matter","rule":"R1","candidates":{"front_matter":"n/a","evidence_card":"no","evidence_card_quoted":false},"meaning":"the paper reports no trading or execution result"},"results":{"status":"available","evaluation_type":"forecast-accuracy","findings":[{"finding":"TimeGAN achieves the lowest Maximum Mean Discrepancy (MMD) of 1.84e-3, indicating the best overall realism among the models.","kind":"statistical","quote":"TimeGAN attained the lowest MMD: 1.84e-3, average over 5 seeds"},{"finding":"In portfolio optimization, TimeGAN’s synthetic data yields a Sharpe ratio difference of -0.03 and a weight cosine similarity of 0.91, closest to real‑data performance.","kind":"performance","quote":"TimeGAN achieves the closest replication of the real-data portfolio, with a cosine similarity of 0.91 and a Sharpe difference of only -0.03."},{"finding":"For volatility forecasting, TimeGAN reduces RMSE to 0.103 and improves directional accuracy to 63%, outperforming ARIMA‑GARCH and VAE.","kind":"predictive-accuracy","quote":"TimeGAN again achieves the best alignment with real-data volatility, reducing RMSE by 17% relative to ARIMA–GARCH and improving directional accuracy to 63%."},{"finding":"ARIMA‑GARCH captures linear trends and conditional volatility but fails to reproduce nonlinear dynamics.","kind":"negative","quote":"ARIMA‑GARCH captures linear trends and conditional volatility but fails to reproduce nonlinear dynamics"},{"finding":"VAEs produce smooth trajectories that underestimate extreme events.","kind":"negative","quote":"VAEs produce smooth trajectories that underestimate extreme events"}],"reported_performance":null,"out_of_sample":{"method":"holdout","headline_out_of_sample":null}},"independent_evidence":{"status":"none_found","supports":[],"contradicts":[],"qualifies":[],"compared_with":55,"citations":{"cited_by_in_corpus":2,"by_independent_authors":1,"how_cited":{"used_as_baseline":0,"uses_method_or_data":0,"extends":0,"discusses_findings":0,"mentions":1,"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":"2026-11-14T17:18:02Z"},"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/"}