{"paper_id":"2410.18897","title":"Generation of synthetic financial time series by diffusion models","authors":["Tomonori Takahashi","Takayuki Mizuno"],"paper_date":"2024-10-24","paper_type":"empirical","paper_url":"https://arxiv.org/pdf/2410.18897v1","page_url":"https://thequant.space/flowcharts/generation-of-synthetic-financial-time-series-by-diffusion-m/","scores":{"empirical_rigor":5,"math_complexity":7,"quadrant":"Holy Grail"},"proposes":{"status":"available","statement":"The paper proposes using denoising diffusion probabilistic models (DDPMs) combined with wavelet transformation to generate synthetic financial time series.","quote":"We alternatively propose utilizing diffusion models, specifically denoising diffusion probabilistic models (DDPMs), to generate synthetic financial time series. This approach employs wavelet transformation to convert multiple time series (into images), such as stock prices, trading volumes, and spreads.","summary":"Despite its practical significance, generating realistic synthetic financial time series is challenging due to statistical properties known as stylized facts, such as fat tails, volatility clustering, and seasonality patterns. Various generative models, including generative adversarial networks…","strategy":null},"assumptions":{"status":"available","items":[{"assumption":"The methodology assumes simultaneous generation of three synchronized observed financial time series: prices, bid/ask spreads, and trading volumes.","type":"data","stated":"explicit","quote":"In the subsequent description, we assume the simultaneous generation of three synchronized observed ﬁnancial time series: prices, bid/ask spreads, and trading volumes.","support":"full"},{"assumption":"Time series are expanded by mirror expansions at both ends to achieve a length of 2n before wavelet transformation.","type":"model","stated":"explicit","quote":"Initially, because sequence data with size 2n simplify the subsequent discrete wavelet transformation, we expand the time series by mirror expansions at both ends to align the length of the time series to 2n.","support":"full"},{"assumption":"The Haar wavelet is used as the mother wavelet for the discrete wavelet transformation.","type":"model","stated":"explicit","quote":"For the transformation, we employed the Haar wavelet as the mother wavelet.","support":"full"},{"assumption":"Power conversion parameters are set to p = 1.5 for price log returns and p = 1.0 for spreads and trading volumes.","type":"model","stated":"explicit","quote":"Parameters: In preprocessing, we adopt power conversion parameter p = 1.5 for price log returns and p = 1.0 for the spreads and trading volumes.","support":"full"},{"assumption":"Winsorization is applied with a level of 10.0, replacing outliers beyond 10σ with 10σ.","type":"model","stated":"explicit","quote":"The win-sorization level is set at 10.0, meaning that outliers beyond 10σ are replaced by 10σ.","support":"full"},{"assumption":"The three time series are treated as three channels of an RGB image for generation.","type":"model","stated":"explicit","quote":"To leverage its advantages (excellent image generation), we convert the time series data into images to train the model. After training, the model generates images, which we then convert back into the original time series data.","support":"full"}]},"datasets":{"status":"available","items":[{"name":"AAPL minute-based stock prices, spreads, and trading volumes","provider":"Refinitiv (LSEG)","kind":"market-prices","access":"commercial","period":"January 2005 to December 2014","universe":"AAPL.O on NASDAQ","quote":"Data: In our experiment, we selected minute-based stock prices, spreads, and trading volumes of AAPL.O traded on NASDAQ from January 2005 to December 2014. We purchased the data from Reﬁnitiv (LSEG), a ﬁnancial data provider.","support":"full","known_id":"refinitiv"}],"known":[{"id":"refinitiv","name":"Refinitiv / LSEG (Datastream, Eikon, Tick History)","provider":"LSEG","kind":"market-prices","access":"commercial","quote":"We purchased the data from Reﬁnitiv (LSEG), a ﬁnancial data provider.","mentions":1,"used_in":1}],"sample":{"start":null,"end":null,"market":null,"asset_class":null,"frequency":null},"data_resolution":"intraday-bars"},"transaction_costs":{"status":"not_applicable","modeled":"n/a","headline_net_of_costs":null,"detail":null,"quote":null,"source":"front_matter","rule":"R3","candidates":{"front_matter":"n/a","evidence_card":"n/a","evidence_card_quoted":false},"meaning":"the paper reports no trading or execution result"},"results":{"status":"available","evaluation_type":"unclear","findings":[{"finding":"The DDPM with wavelet imaging more effectively replicates financial stylized facts—including fat tails, slow decay of autocorrelations (volatility clustering), intraday seasonality, and cross‑correlation structures—than TimeGAN, QuantGAN, and a simple DDPM wit","kind":"statistical","quote":"The DDPM and wavelet imaging approaches more eﬀectively replicated characteristics commonly observed in ﬁnancial time series, such as the fat tails, the slow decay of autocorrelations including volatility clustering, the intraday seasonal-ity patterns, and the cross correlation coeﬃcients among time series, compared to TimeGAN and QuantGAN approaches and the simple application of DDPM without wavelet imaging.","support":"full"}],"reported_performance":null,"out_of_sample":{"method":"none","headline_out_of_sample":null}},"independent_evidence":{"status":"none_found","supports":[],"contradicts":[],"qualifies":[],"compared_with":5,"citations":{"cited_by_in_corpus":3,"by_independent_authors":3,"how_cited":{"used_as_baseline":0,"uses_method_or_data":0,"extends":0,"discusses_findings":0,"mentions":3,"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":2,"extracted_at":"2026-11-14T19:54:23Z"},"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/"}