{"paper_id":"2408.01387","title":"NeuralBeta: Estimating Beta Using Deep Learning","authors":["Yuxin Liu","Jimin Lin","Achintya Gopal"],"paper_date":"2024-08-02","paper_type":"empirical","paper_url":"https://arxiv.org/pdf/2408.01387v2","page_url":"https://thequant.space/flowcharts/neuralbeta-estimating-beta-using-deep-learning/","scores":{"empirical_rigor":8,"math_complexity":6.5,"quadrant":"Holy Grail"},"proposes":{"status":"available","statement":"The paper proposes NeuralBeta, a deep‑learning model for estimating dynamic beta, including an interpretable version with a weighted‑linear‑regression output layer.","quote":"we have developed a novel method using neural networks called NeuralBeta, which is capable of handling both univariate and multivariate scenarios and tracking the dynamic behavior of β.","summary":"Traditional approaches to estimating beta in finance often involve rigid assumptions and fail to adequately capture beta dynamics, limiting their effectiveness in use cases like hedging. To address these limitations, we have developed a novel method using neural networks called NeuralBeta, which is…","strategy":null},"assumptions":{"status":"available","items":[{"assumption":"The data used for ordinary linear regression satisfy the Gauss‑Markov assumptions.","type":"statistical","stated":"explicit","quote":"if the data 𝐷0,τ+1 satisfies the Gauss‑Markov assumptions, then the optimal functional form of β is given by"},{"assumption":"The true beta coefficient is not assumed to be time‑homogeneous.","type":"model","stated":"explicit","quote":"we don’t presume the ground truth coefficient β to be time‑homogeneous, so the coefficient is expressed in a functional form as in equation (2)"},{"assumption":"A risk measure L (e.g., expected quadratic loss or negative log‑likelihood) is used to define the hedging objective.","type":"model","stated":"explicit","quote":"is some risk measure, such as expected quadratic loss or negative log‑likelihood."}]},"datasets":{"status":"available","items":[{"name":"Synthetic sinusoidal beta datasets","provider":null,"kind":"synthetic","access":"unknown","period":null,"universe":null,"quote":"The parameters are chosen such that each sample contains at least one half period of the sinusoidal wave."},{"name":"Market data","provider":null,"kind":"market-prices","access":"unknown","period":null,"universe":null,"quote":"We conducted extensive experiments on both synthetic and market data, demonstrating NeuralBeta’s superior performance compared to benchmark methods across various scenarios"}],"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":"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":"NeuralBeta significantly outperforms rolling OLS and rolling WLS on synthetic constant, stepwise, and cyclical beta scenarios.","kind":"performance","quote":"In Table 1, we can see in all three cases of synthetic β, NeuralBeta significantly outperforms OLS and WLS."},{"finding":"The interpretable NeuralBeta (with attention) achieves almost identical performance to the non‑interpretable version.","kind":"performance","quote":"the performance of interpretable NeuralBeta and non‑interpretable ones are almost identical, and interpretable NeuralBeta with attention mechanism outperforms except for the cyclical β case."},{"finding":"NeuralBeta automatically learns weighting schemes (e.g., half‑life of 2) that match the best‑performing WLS strategies for stepwise beta.","kind":"predictive-accuracy","quote":"NeuralBeta automatically identifies and applies this strategy."},{"finding":"NeuralBeta provides smoother and more accurate estimations than WLS for cyclical beta, achieving the largest improvement over OLS when the sinusoid period is longer.","kind":"performance","quote":"NeuralBeta outperforms both OLS and WLS. While WLS, with a half‑life of 1, can track the sine wave’s trend, NeuralBeta provides smoother and more accurate estimations, while maintaining the same level of responsiveness."}],"reported_performance":null,"out_of_sample":{"method":"holdout","headline_out_of_sample":null}},"independent_evidence":{"status":"none_found","supports":[],"contradicts":[],"qualifies":[],"compared_with":19,"citations":{"cited_by_in_corpus":1,"by_independent_authors":1,"how_cited":{"used_as_baseline":0,"uses_method_or_data":0,"extends":0,"discusses_findings":0,"mentions":1,"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":"2026-11-14T18:04:55Z"},"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/"}