A Controlled Comparison of Deep Learning Architectures for Multi-Horizon Financial Forecasting: Evidence from 918 Experiments

arXiv:2603.16886 · q-fin.ST, cs.LG, q-fin.GN · Submitted 2026-02-27 · Read on arXiv

q-fin.ST, cs.LG, q-fin.GN

Submitted: 2026-02-27

Updated: 2026-10-30

Terminology

• Accuracy — 1× in this paper · explained in 1 episode
• Benchmark — 1× in this paper · explained in 2 episodes
• Inductive Bias — 1× in this paper · explained in 3 episodes
• Transformer — 1× in this paper · explained in 2 episodes

Related papers

• Physics-Informed Singular-Value Learning for Cross-Covariances Forecasting in Financial Markets
• Residual Learning in Empirical Asset Pricing
• Multiscaling in the Rough Bergomi Model: A Tale of Tails
• Fast Approximate Modular Inference for Option Price Prediction
• Financial Anomaly Detection for the Canadian Market
• Adapting Altman's bankruptcy prediction model to the compositional data methodology