Paper: arXiv 2610.09654

Authors: Aashish Bohra, Lokendra Vishwakarm

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.

Complexity vs Empirical Score

  • Math Complexity: 7.5/10
  • Empirical Rigor: 8.0/10
  • Quadrant: Holy Grail — high math complexity, high empirical rigor

Why this score: This paper presents a highly novel approach to financial forecasting using dynamic spectral trajectories and a sophisticated deep learning architecture. It demonstrates strong empirical rigor through extensive evaluation against multiple baselines and a robust hold-out period. The mathematical underpinnings, including causality proofs and complexity analysis, are significant.

Research Flowchart

  flowchart TD
    A[Research Goal: Causal Multi-Horizon Financial Forecasting] --> B{Key Methodology: DSTNet}

    B --> C[Data/Inputs: 7 Equity Indices, Gold, 7 Trailing Technical Indicators (20-day lookback)]
    C --> D[Computational Processes: One-sided Morlet Filter Bank, Dynamic Spectral Trajectory, Factorized Scale-Temporal Spectral Transformer, CNN-BiLSTM, Horizon-specific Gates]
    D --> E[Outcomes: 1, 3, 5, 10 Day Forecasts]

    E --> F{Key Findings: DSTNet consistently outperforms baselines and persistence by 0.7-4.5% MAE/MAPE. No equity-timing advantage found.}