{"paper_id":"2609.23378","title":"Leaky-integrator reconstruction: taming error accumulation in recursive differenced time-series forecasting","authors":["Zijiang Yang"],"paper_date":"2026-09-23","paper_type":"empirical","paper_url":"https://arxiv.org/pdf/2609.23378","page_url":"https://thequant.space/flowcharts/leaky-integrator-reconstruction-taming-error-accumulation-in-recursive/","scores":{"empirical_rigor":8,"math_complexity":6.5,"quadrant":"Holy Grail"},"proposes":{"status":"available","statement":"The paper proposes a leaky‑integrator reconstruction, a training‑free fix that moves the pole inside the unit circle to bound error accumulation in recursive differenced forecasting.","quote":"We then introduce leaky-integrator reconstruction, a training-free fix that moves the pole inside the unit circle with H(z) = 1/(1 - gamma z^-1), gamma < 1, bounding the accumulation of the model's own increment errors.","summary":"Recursive differenced forecasting, the standard remedy for non-stationarity, predicts one-step changes and integrates them by cumulative summation. We show that this reconstruction is a discrete integrator with a pole on the unit circle, so the biased increment errors of a learned nonlinear model…","strategy":null},"assumptions":{"status":"available","items":[{"assumption":"Recursive differenced forecasting reconstructs levels with a unit‑pole integrator (1/(1‑z⁻¹)).","type":"model","stated":"explicit","quote":"the reconstruction is a discrete integrator 1/(1 −z−1) with a pole at z = 1"},{"assumption":"The learned one‑step increment predictor is biased and its errors are correlated.","type":"statistical","stated":"explicit","quote":"a learned model’s biased, correlated increment errors are integrated without bound"},{"assumption":"The evaluation uses normalised MAE (absolute error divided by training standard deviation).","type":"statistical","stated":"explicit","quote":"We report normalised MAE: the absolute level error in units of the training standard deviation, |ˆyt+k − yt+k|/σtrain"},{"assumption":"Stable linear‑AR models are already stable and should remain unchanged (γ=1).","type":"model","stated":"explicit","quote":"Linear‑AR is stable and, correctly, left at γ=1"},{"assumption":"The super‑√k diagnostic can identify when the fix should be applied.","type":"other","stated":"explicit","quote":"gated by the super‑√k diagnostic on a validation split so that stable or joint models are left at γ=1"}]},"datasets":{"status":"none_found","items":[],"known":[],"sample":{"start":null,"end":null,"market":null,"asset_class":null,"frequency":null},"data_resolution":"intraday-bars"},"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":"Recursive differenced forecasting diverges for all nonlinear architectures, reaching up to 3.8 normalised MAE at horizon 336.","kind":"performance","quote":"on twenty datasets every recursively rolled‑out non‑linear architecture diverges (up to 3.8 normalised MAE at H=336)"},{"finding":"Leaky‑integrator reconstruction with fixed γ=0.9 reduces 336‑step normalised MAE by 43–75% across the seven diverging architectures, bringing errors to 0.87–0.97.","kind":"performance","quote":"a single fixed pole γ=0.9 pulls all of them back to ≈0.87–0.97 nMAE, a 43–75% error reduction"},{"finding":"The gain from the leaky integrator grows with horizon, e.g., ~3% at H=24, 23% at H=96, 37% at H=192, and 51% at H=336.","kind":"performance","quote":"the mean gain over seven diverging architectures and twenty datasets growing from ~3% at H=24 to 23% at H=96, 37% at H=192 and 51% at H=336"},{"finding":"Linear‑AR models remain unchanged (γ=1) and retain their baseline normalised MAE of 0.80.","kind":"performance","quote":"Linear‑AR is stable and, correctly, left at γ=1"},{"finding":"The leaky integrator is inert for joint predictors that never integrate increments and can hurt stable recursions (linear‑AR +9.3%).","kind":"negative","quote":"it is inert for joint predictors, which never integrate increments, and hurts stable recursions (linear‑AR +9.3%)"}],"reported_performance":null,"out_of_sample":{"method":"holdout","headline_out_of_sample":null}},"independent_evidence":{"status":"none_found","supports":[],"contradicts":[],"qualifies":[],"compared_with":24,"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":"2026-11-14T17:04:50Z"},"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/"}