{"paper_id":"2509.11844","title":"ProteuS: A Generative Approach for Simulating Concept Drift in Financial Markets","authors":["Andrés L. Suárez-Cetrulo","Alejandro Cervantes","David Quintana"],"paper_date":"2025-08-30","paper_type":"empirical","paper_url":"https://arxiv.org/pdf/2509.11844v1","page_url":"https://thequant.space/flowcharts/proteus-a-generative-approach-for-simulating-concept-drift/","scores":{"empirical_rigor":8,"math_complexity":6.5,"quadrant":"Holy Grail"},"proposes":{"status":"available","statement":"The paper proposes ProteuS, a novel framework for generating semi‑synthetic financial time series with predefined structural breaks and known ground truth for concept drift.","quote":"this paper introduces the ProteuS framework, a novel methodology for generating semi-synthetic financial time series with pre-defined structural breaks.","summary":"Financial markets are complex, non-stationary systems where the underlying data distributions can shift over time, a phenomenon known as regime changes, as well as concept drift in the machine learning literature. These shifts, often triggered by major economic events, pose a significant challenge…","strategy":null},"assumptions":{"status":"available","items":[{"assumption":"ARMA‑GARCH models can capture the distinct statistical properties of different market regimes.","type":"model","stated":"explicit","quote":"Using ProteuS, we have fitted ARMA-GARCH models to real-world ETF data and were able to capture the distinct statistical properties of different market regimes."},{"assumption":"The semi‑synthetic data generated by fitting ARMA‑GARCH to ETF data and simulating transitions is realistic enough for evaluating drift detection.","type":"data","stated":"explicit","quote":"Our methodology involves fitting ARMA-GARCH models to real-world ETF data to capture distinct market regimes, and then simulating realistic, gradual, and abrupt transitions between them."},{"assumption":"Technical indicators (including added 5‑minute moving averages) are sufficient to represent the feature space for drift detection experiments.","type":"data","stated":"explicit","quote":"We have augmented this base set with the inclusion of additional short-term (5-minute) moving averages."},{"assumption":"A known ground truth of when concept changes occur can be established using the generated datasets.","type":"other","stated":"explicit","quote":"By using these datasets, we establish a known ground truth for when concept changes occur."}]},"datasets":{"status":"available","items":[{"name":"ProteuS generated synthetic data streams","provider":null,"kind":"synthetic","access":"unknown","period":null,"universe":null,"quote":"we used ProteuS to generate 30 unique semi-synthetic data streams, each containing 1.5 million data points."}],"known":[],"sample":{"start":null,"end":null,"market":null,"asset_class":null,"frequency":null},"data_resolution":"unclear"},"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":"none","findings":[{"finding":"The generated datasets show a high degree of overlap between different market states, making regime identification non‑trivial.","kind":"statistical","quote":"Our analysis of the generated datasets revealed a high degree of overlap between the different market states, confirming that the task of identifying and adapting to these regimes is non-trivial."},{"finding":"The algorithm fails to find distinct clusters for the four ground‑truth states, highlighting classification difficulty.","kind":"negative","quote":"The inability of the algorithm to find distinct clusters corresponding to the four ground truth states highlights the high degree of overlap and the difficulty of the classification task."},{"finding":"ProteuS successfully creates 30 semi‑synthetic streams of 1.5 million points each, providing a controlled benchmark for drift detection.","kind":"other","quote":"we used ProteuS to generate 30 unique semi-synthetic data streams, each containing 1.5 million data points."}],"reported_performance":null,"out_of_sample":{"method":"none","headline_out_of_sample":null}},"independent_evidence":{"status":"none_found","supports":[],"contradicts":[],"qualifies":[],"compared_with":16,"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":true,"code_url":"https://github.com/cetrulin/regime-switching-series-generator","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":"2026-11-14T16:38:19Z"},"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/"}