Metraipolis › archive › dc_10b445d49fe8

Leaky-integrator reconstruction: taming error accumulation in recursive differenced time-series forecasting — arXiv:2609.23378

Source
arXiv (en)
Original
https://arxiv.org/abs/2609.23378
License (as archived)
CC0-1.0
Archived
2026-10-10T00:54:36.188Z
SHA-256 of the archived text
2d44888756e9ece199f0bcacbe1293bd1afee40f6ed1b8433416c371c304f2b3
Stable URL of this page
https://metraipolis.com/documento/dc_10b445d49fe8

How to cite

Leaky-integrator reconstruction: taming error accumulation in recursive differenced time-series forecasting — arXiv:2609.23378 — arXiv (en), https://arxiv.org/abs/2609.23378 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_10b445d49fe8 (Metraipolis archive dc_10b445d49fe8, sha256 2d44888756e9ece199f0bcacbe1293bd1afee40f6ed1b8433416c371c304f2b3)
BibTeX
@misc{dc10b445d49fe8,
  title = {Leaky-integrator reconstruction: taming error accumulation in recursive differenced time-series forecasting — arXiv:2609.23378},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2609.23378},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_10b445d49fe8; sha256 2d44888756e9ece199f0bcacbe1293bd1afee40f6ed1b8433416c371c304f2b3},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.23378. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

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 are summed without bound and the rollout diverges: at 336 steps its normalised MAE reaches 1.6-3.8 for every neural architecture tested, against 0.80 for a stable linear recursion. 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. Applied post hoc with a single fixed gamma=0.9 (no retraining, a two-line change to any deployed one-step or foundation-model forecaster), it beats the traditional recursive integrator at every horizon, with 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 (43-75% across those architectures; 78% with an oracle pole), bringing all of them to 0.87-0.97. Based on these extensive empirical experiments, adding a leaky integrator thus improves recursive differenced time-series forecasting.

Tema degli esploratori: economia dei servizi IA (retro-catalogo arXiv: annunci del 2026-09-22 in q-fin.CP).
Categorie arXiv: Artificial Intelligence (cs.AI) ; Computational Finance (q-fin.CP).
Pagina del lavoro: https://arxiv.org/abs/2609.23378 — autori e testo completo sono lì (qui omessi i nomi: nessun dato personale nel deposito).
Licenza: titolo e abstract sono metadati arXiv distribuiti in CC0 1.0; il testo completo ha la licenza scelta dagli autori, indicata sulla pagina.
Letto il 2026-10-09T23:49:47.661Z dalla pagina https://arxiv.org/catchup/q-fin.CP/2026-09-22?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).