{"paper_id":"2403.14695","title":"Chain-structured neural architecture search for financial time series forecasting","authors":["Denis Levchenko","Efstratios Rappos","Shabnam Ataee","Biagio Nigro","Stephan Robert-Nicoud"],"paper_date":"2024-03-15","paper_type":"methodological","paper_url":"https://arxiv.org/pdf/2403.14695v2","page_url":"https://thequant.space/flowcharts/chain-structured-neural-architecture-search-for-financial-ti/","scores":{"empirical_rigor":7,"math_complexity":4.5,"quadrant":"Street Traders"},"proposes":{"status":"available","statement":"The paper evaluates three NAS strategies on chain-structured search spaces for financial time series forecasting and reports that Bayesian optimization and hyperband perform best, with RNN and 1D CNN yielding the top architectures.","quote":"We find Bayesian optimization and the hyperband method performing best among the strategies, and RNN and 1D CNN best among the architectures, but all methods were very close to each other with a high variance due to the difficulty of working with financial datasets.","summary":"Neural architecture search (NAS) emerged as a way to automatically optimize neural networks for a specific task and dataset. Despite an abundance of research on NAS for images and natural language applications, similar studies for time series data are lacking. Among NAS search spaces…","strategy":null},"assumptions":{"status":"available","items":[{"assumption":"Data are split chronologically with 70% training, 20% validation, and 10% test.","type":"data","stated":"explicit","quote":"The data was split into (in historical order): first 70% for training, the following 20% for validation used in the evaluation during optimization runs, and the last 10% for final testing on unseen data."},{"assumption":"Financial time series are noisy and non‑stationary, leading to low signal‑to‑noise ratios.","type":"market","stated":"explicit","quote":"The problem is challenging due to the inherently noisy and non-stationary nature of financial time series."},{"assumption":"Binary classification performance is evaluated primarily with AUC, balanced accuracy and F1, assuming these metrics reflect predictive usefulness.","type":"statistical","stated":"explicit","quote":"We mostly focused on AUC as we found it to be the hardest metric to optimize, but we paid close attention to bACC and F1 too and report…"}]},"datasets":{"status":"available","items":[{"name":"Germany dataset","provider":null,"kind":"market-prices","access":"unknown","period":null,"universe":null,"quote":"Our best results were obtained on the Germany dataset, with time-derived redundant features removed but no PCA applied."},{"name":"Japan dataset","provider":null,"kind":"market-prices","access":"unknown","period":null,"universe":null,"quote":"For the Japan dataset, the redundant time-derived features removed and PCA applied retaining most of the information in the dataset while further cutting the number of features approximately in half gave the best results."},{"name":"US dataset","provider":null,"kind":"market-prices","access":"unknown","period":null,"universe":null,"quote":"The US dataset was most challenging for us. Although tuned 1d CNNs gave an AUC score of 0.6 ± 0.02 on validation data (not used for training)"}],"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":"forecast-accuracy","findings":[{"finding":"On the Germany dataset, an LSTM tuned by hyperband achieved an average AUC of 0.56 ± 0.05 and balanced accuracy of 0.54 ± 0.04 over 50 test runs.","kind":"performance","quote":"The LSTM with parameters selected by the hyperband method applied to the unseen test data achieved an AUC score of 0.56 on average over 50 test runs starting from different random seeds (see Section 5), with 0.05 standard deviation. The same architecture showed balanced accuracy score of 0.54 ± 0.04."},{"finding":"A 1D CNN found by Bayesian optimization on the Germany dataset obtained AUC 0.54 ± 0.05 and a high F1 of 0.6 ± 0.06.","kind":"performance","quote":"1D CNNs were also able to achieve good performance on this dataset, with the best architecture provided by Bayesian optimization giving an AUC score of 0.54 ± 0.05 and a high F1 score of 0.6 ± 0.06."},{"finding":"For the Japan dataset, the best model was a 1D CNN from Bayesian optimization with AUC 0.54 ± 0.03 and F1 0.65 ± 0.02.","kind":"performance","quote":"The best performing architecture was a 1D CNN coming from Bayesian optimization, achieving an AUC score of 0.54 ± 0.03 over 50 test runs, with a high F1 score of 0.65 ± 0.02."},{"finding":"On the US dataset, no architecture achieved average test AUC above 0.5, indicating poor predictive performance.","kind":"negative","quote":"on average no architecture could achieve AUC over 0.5 on the test dataset."},{"finding":"Overall, Bayesian optimization and hyperband outperformed reinforcement learning, while LSTMs and 1D CNNs consistently beat simple feed‑forward networks across datasets.","kind":"performance","quote":"For search strategy, overall the hyperband method and Bayesian optimization showed better performance than the reinforcement learning based approach, but all three were very close to each other."}],"reported_performance":null,"out_of_sample":{"method":"holdout","headline_out_of_sample":null}},"independent_evidence":{"status":"none_found","supports":[],"contradicts":[],"qualifies":[],"compared_with":25,"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":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":"2026-11-14T18:13:14Z"},"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/"}