{"paper_id":"2402.06635","title":"Large (and Deep) Factor Models","authors":["Bryan Kelly","Boris Kuznetsov","Semyon Malamud","Teng Andrea Xu","Yuan Zhang"],"paper_date":"2024-01-20","paper_type":"empirical","paper_url":"https://arxiv.org/pdf/2402.06635v1","page_url":"https://thequant.space/flowcharts/large-and-deep-factor-models/","scores":{"empirical_rigor":6,"math_complexity":9,"quadrant":"Holy Grail"},"proposes":{"status":"available","statement":"The paper proposes a deep neural network framework that decomposes the stochastic discount factor into nonlinear characteristic discovery and a pricing rule captured by the Portfolio Tangent Kernel (PTK).","quote":"We show that a deep neural network (DNN) trained to construct a stochastic discount factor (SDF) admits a sharp additive decomposition that separates nonlinear characteristic discovery from the pricing rule that aggregates them.","summary":"We open up the black box behind Deep Learning for portfolio optimization and prove that a sufficiently wide and arbitrarily deep neural network (DNN) trained to maximize the Sharpe ratio of the Stochastic Discount Factor (SDF) is equivalent to a large factor model (LFM): A linear factor pricing…","strategy":null},"assumptions":{"status":"available","items":[{"assumption":"Factor returns follow Ft+1 = µP + Σ1/2 P Xt with i.i.d. components and finite eighth moments.","type":"model","stated":"explicit","quote":"Suppose Ft+1 = µP + Σ1/2 P Xt, where Xt has i.i.d. components with E[X8+ε i,t ] < ∞."},{"assumption":"After a large number of gradient‑descent steps the PTK stabilizes and is non‑degenerate.","type":"model","stated":"explicit","quote":"Suppose that, after s∗gradient descent steps, the PTK stabilizes in the sense that"},{"assumption":"The data generating process is stationary enough that rolling‑window estimation of the SDF is valid.","type":"data","stated":"implied","quote":"We use these normalized factor returns to estimate a candidate stochastic discount factor as the return on a Markowitz portfolio computed over a rolling window of T months with ridge penalty z."}]},"datasets":{"status":"available","items":[{"name":"JKP characteristics dataset","provider":"Jensen, Kelly, and Pedersen (2023)","kind":"fundamentals","access":"public","period":"1963–2024","universe":"NYSE, AMEX, and NASDAQ stocks with CRSP share codes 10–12","quote":"We use monthly data on 153 stock characteristics for U.S. equities over the period 1963–2024 from (Jensen, Kelly, and Pedersen, 2023, hereafter JKP). This open‑source dataset collects a comprehensive set of firm‑level characteristics","known_id":"jkp"},{"name":"Random‑feature ReLU benchmark","provider":null,"kind":"benchmark","access":"public","period":null,"universe":null,"quote":"As a benchmark, we construct random‑feature‑based SDFs following Didisheim et al. (2024). Specifically, we define St = ReLU(Xtγ) ∈RNt×P, where γ = (γj,k) ∈Rd×P with P = 25,000 and γj,k ∼N(0, 1) independently."}],"known":[{"id":"jkp","name":"JKP Global Factor Data","provider":"Jensen, Kelly and Pedersen","kind":"market-prices","access":"public","quote":"Data and Random Feature Portfolios We use monthly data on 153 stock characteristics for U.S. equities over the period 1963–2024 from (Jensen, Kelly, and Pedersen, 2023, hereafter JKP).","mentions":2,"used_in":1},{"id":"crsp","name":"CRSP","provider":"Center for Research in Security Prices","kind":"market-prices","access":"commercial","quote":"This open-source dataset collects a comprehensive set of firm-level characteristics drawn from the asset 27 pricing literature.5 The universe consists of NYSE, AMEX, and NASDAQ stocks with CRSP share codes 10–12.","mentions":1,"used_in":1}],"sample":{"start":null,"end":null,"market":null,"asset_class":null,"frequency":null},"data_resolution":"weekly-monthly"},"transaction_costs":{"status":"available","modeled":"unclear","headline_net_of_costs":null,"detail":null,"quote":null,"source":"conflict","rule":"R4b","candidates":{"front_matter":"no","evidence_card":"n/a","evidence_card_quoted":false},"conflict_kind":"applicability","meaning":"the platform classifier and the paper-text extraction disagree, so no value is asserted"},"results":{"status":"available","evaluation_type":"trading-pnl","findings":[{"finding":"The PTK‑SDF achieves substantially higher Sharpe ratios than both the trained DNN and the random‑feature ReLU benchmark across all size groups.","kind":"performance","quote":"Figure 4 confirms this prediction. The PTK‑SDF delivers substantially higher Sharpe ratios than both benchmarks across all size groups."},{"finding":"A depth‑2, width‑256 DNN attains a Sharpe ratio of 3.8, slightly exceeding the ReLU benchmark’s 3.7.","kind":"performance","quote":"For example, the depth‑2, width‑256 network attains a Sharpe ratio of 3.8, slightly exceeding the ReLU benchmark (3.7)."},{"finding":"Moderately complex DNNs (≈100,000 parameters) discover priced nonlinear characteristics not spanned by random features, yielding a significant positive alpha in spanning tests.","kind":"statistical","quote":"The results strongly support the feature‑learning hypothesis. For moderately complex deep architectures"},{"finding":"Shallow networks (depth = 1) perform poorly and substantially underperform the random‑feature benchmark.","kind":"negative","quote":"Shallow networks (D = 1) perform poorly and substantially underperform the random feature benchmark."},{"finding":"When the PTK is used, it isolates the economically relevant component of feature learning and translates it into a disciplined pricing rule, yielding tradable SDFs tightly linked to long‑horizon consumption risk.","kind":"theoretical","quote":"Overall, PTK isolates the economically relevant component of feature learning and translates it into a disciplined pricing rule, yielding tradable SDFs that are tightly linked to long‑horizon consumption risk."}],"reported_performance":null,"out_of_sample":{"method":"walk-forward","headline_out_of_sample":null}},"independent_evidence":{"status":"none_found","supports":[],"contradicts":[],"qualifies":[],"compared_with":39,"citations":{"cited_by_in_corpus":2,"by_independent_authors":2,"how_cited":{"used_as_baseline":0,"uses_method_or_data":0,"extends":0,"discusses_findings":0,"mentions":2,"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":"2026-11-14T18:16:14Z"},"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/"}