Newest papers

DSTNet: Dynamic Spectral Trajectory Network for Causal Multi-Horizon Financial Forecasting

Aashish Bohra, Lokendra Vishwakarm · 2026 · arXiv (Cornell University)

DOI: 10.48550/arxiv.2610.09654 · cited by 0 works (OpenAlex) · open-access PDF

Russian

Ультрафиолетовые финансовые прогнозы, как правило, используют трансформацию только для денонсации или сокращения ее до одного спектрального снижения при прогнозируемом происхождении, а конволуция, которая производит коэффициенты, обычно двусторонняя, поэтому она может прочесть за пределы прогнозируемого происхождения. DSTNet вместо этого сохраняет недавнюю эволюцию величин фильтра-банка в качестве причинно-динамической спектральной траектории, построенную из семи последующих технических индикаторов в течение двадцатидневного обзора с любой односторонней системой фильтра Morlet-derived и явным выгоранием для левостороннего транзиента. Факторизированный Scale-Temporal Spectral Transformer

Machine translation (Meta M2M100); the English original follows.

Abstract

Wavelet-based financial forecasters typically use the transform only to denoise, or reduce it to a single spectral snapshot at the forecast origin, and the convolution that produces the coefficients is usually bilateral, so it can read past the forecast origin. DSTNet instead retains the recent evolution of filter-bank magnitudes as a causal Dynamic Spectral Trajectory, built from seven trailing technical indicators over a twenty-day lookback with a one-sided Morlet-derived filter bank and an explicit burn-in for the left-boundary transient. A factorized Scale-Temporal Spectral Transformer attends along the time and filter-bank axes separately, a learned gate fuses the spectral branch with a CNN-BiLSTM, and horizon-specific gates emit one, three, five, and ten day forecasts in a single pass. We evaluate seven equity indices and gold under a common expanding-window protocol and an untouched one-year hold-out, against nine learned baselines and a random-walk persistence benchmark. Under MAE and MAPE, persistence is the strongest of the ten fixed competitors in 29 of the 32 series-horizon cells and DSTNet is the only model below it in every cell, by 0.7 to 0.9 percent at one day and 3.4 to 4.5 percent at ten days. At one day, paired testing favours DSTNet against the weaker learned baselines but is inconclusive against persistence and the strongest learned forecasters. A downstream allocation diagnostic does not support an equity-timing advantage on any of the seven indices.

Read in: English · Spanish · Chinese · Hindi · Arabic · Portuguese · French · German · Japanese · Russian · Korean · Indonesian · Vietnamese

Topics: Econometrics, Snapshot (computer storage), Computer science, Equity (law), Spectral analysis, Mathematics

Sources: OpenAlex (CC0), arXiv, Crossref. Agent Exchange