{"paper_id":"2601.07131","title":"The Limits of Complexity: Why Feature Engineering Beats Deep Learning in Investor Flow Prediction","authors":["Sungwoo Kang"],"paper_date":"2026-01-12","paper_type":"empirical","paper_url":"https://arxiv.org/pdf/2601.07131v1","page_url":"https://thequant.space/flowcharts/the-limits-of-complexity-why-feature-engineering-beats-deep/","scores":{"empirical_rigor":9,"math_complexity":7.5,"quadrant":"Holy Grail"},"proposes":{"status":"available","statement":"The paper proposes that simple market‑cap‑normalized flow (Matched Filter) linear models outperform advanced ICA‑Wavelet‑LSTM pipelines for investor flow prediction.","quote":"We bridge these literatures by testing whether advanced signal processing (ICA, Wavelets) and deep learning (LSTM) can extract additional predictive value from investor flows beyond simple normalization.","summary":"The application of machine learning to financial prediction has accelerated dramatically, yet the conditions under which complex models outperform simple alternatives remain poorly understood. This paper investigates whether advanced signal processing and deep learning techniques can extract…","strategy":"Simple Momentum (Linear) with Matched Filter normalization"},"assumptions":{"status":"available","items":[{"assumption":"Transaction costs are 10 basis points per round‑trip trade.","type":"execution","stated":"explicit","quote":"Sharpe Ratio (annualized, net of 10bp transaction costs)"},{"assumption":"Trading strategies are evaluated on daily data with no slippage beyond the 10bp cost.","type":"execution","stated":"implied","quote":"Net Sharpe accounts for 10bp round‑trip transaction costs."},{"assumption":"Investor flow data can be normalized by market capitalization to create a scale‑invariant measure.","type":"data","stated":"explicit","quote":"Following the signal processing concept of matched filtering, we normalize flows to create a scale‑invariant measure: Sg i,t = NetBuyg i,t / MarketCapi,t"},{"assumption":"The ICA mixing model assumes observed flows are linear mixtures of independent latent components plus noise.","type":"model","stated":"explicit","quote":"We hypothesize that observed investor flows are mixtures of latent market drivers: Xt = A · St + ϵt"},{"assumption":"The LSTM is trained to predict next‑day returns using mean‑squared error loss.","type":"model","stated":"explicit","quote":"Loss function: Mean Squared Error (MSE) • Optimizer: Adam with learning rate 10−3"}]},"datasets":{"status":"available","items":[{"name":"Korean equities daily trading records","provider":"Korea Exchange","kind":"market-prices","access":"proprietary","period":"2020‑2024","universe":"2,439 Korean equities (KOSPI and KOSDAQ)","quote":"Our dataset comprises daily trading records for all common stocks listed on the Korea Exchange (KOSPI and KOSDAQ) from January 2, 2020 through December 27, 2024."}],"known":[],"sample":{"start":"2020","end":"2024","market":"Korean equities","asset_class":"equities","frequency":"daily"},"data_resolution":"daily"},"transaction_costs":{"status":"available","modeled":"yes","headline_net_of_costs":null,"detail":"10 basis points round‑trip transaction cost applied to reported Sharpe ratios.","quote":"Sharpe Ratio (annualized, net of 10bp transaction costs)","source":"evidence_card","rule":"R2","candidates":{"front_matter":"unclear","evidence_card":"yes","evidence_card_quoted":true},"card_detail":"10 basis points round‑trip transaction cost applied to reported Sharpe ratios.","card_quote":"Sharpe Ratio (annualized, net of 10bp transaction costs)","meaning":"costs are part of the reported headline results"},"results":{"status":"available","evaluation_type":"trading-pnl","findings":[{"finding":"Simple Momentum using Matched Filter normalized flows achieves a Sharpe ratio of 1.30 and cumulative return of 272.6% over five years.","kind":"performance","quote":"Simple Momentum achieves a Sharpe ratio of 1.30 after transaction costs, with cumulative return of 272.6% over 5 years (annualized ∼30%)."},{"finding":"The ICA‑Macro Factor strategy yields a Sharpe ratio of only 0.07 and a negative cumulative return of –5.1%.","kind":"performance","quote":"ICA Macro Factor generates a Sharpe of only 0.07—essentially zero risk‑adjusted return—with a catastrophic 48.9% maximum drawdown and negative cumulative return."},{"finding":"The raw LSTM model collapses to predicting the unconditional mean, with a hit rate of 47.5%, worse than random.","kind":"predictive-accuracy","quote":"The raw LSTM model collapsed to predicting the unconditional mean, achieving a hit rate of 47.5% -- worse than random."},{"finding":"Complexity destroys information: the full ICA‑Wavelet‑LSTM pipeline produces a Sharpe of 0.07 versus 1.30 for the simple linear model.","kind":"negative","quote":"Our findings are unambiguous: complexity fails."},{"finding":"Normalization creates a sufficient statistic for return prediction, making additional processing detrimental.","kind":"theoretical","quote":"The normalized flow may be a sufficient statistic for return prediction—containing all relevant information such that additional processing can only lose rather than gain information."}],"reported_performance":{"strategy":"Simple Momentum (Linear) with Matched Filter normalization","sharpe":1.3,"sharpe_benchmark":null,"annual_return_pct":null,"total_return_pct":272.6,"max_drawdown_pct":19.1},"out_of_sample":{"method":"holdout","headline_out_of_sample":true}},"independent_evidence":{"status":"none_found","supports":[],"contradicts":[],"qualifies":[],"compared_with":43,"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":"2026-11-14T17:16:13Z"},"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/"}