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

• URL: http://arxiv.org/abs/2603.16886v1
• Date: Fri, 27 Feb 2026 03:27:04 GMT
• Title: A Controlled Comparison of Deep Learning Architectures for Multi-Horizon Financial Forecasting: Evidence from 918 Experiments
• Abstract summary: Multi-horizon price forecasting is central to portfolio allocation, risk management, and algorithmic trading.<n>Deep learning architectures have proliferated faster than rigorous financial benchmarks can evaluate them.<n>ModernTCN achieves the best mean rank (1.333) with a 75 percent first-place rate, followed by PatchTST (2..000)
• Score: 0.0
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Multi-horizon price forecasting is central to portfolio allocation, risk management, and algorithmic trading, yet deep learning architectures have proliferated faster than rigorous financial benchmarks can evaluate them. This study provides a controlled comparison of nine architectures (Autoformer, DLinear, iTransformer, LSTM, ModernTCN, N-HiTS, PatchTST, TimesNet, and TimeXer) spanning Transformer, MLP, CNN, and RNN families across cryptocurrency, forex, and equity index markets at 4-hour and 24-hour horizons. A total of 918 experiments were conducted under a strict five-stage protocol including fixed-seed Bayesian hyperparameter optimization, configuration freezing per asset class, multi-seed retraining, uncertainty aggregation, and statistical validation. ModernTCN achieves the best mean rank (1.333) with a 75 percent first-place rate, followed by PatchTST (2.000). Results reveal a clear three-tier ranking structure and show that architecture explains nearly all performance variance, while seed randomness is negligible. Rankings remain stable across horizons despite 2 to 2.5 times error amplification. Directional accuracy remains near 50 percent across all configurations, indicating that MSE-trained models lack directional skill at hourly resolution. The findings highlight the importance of architectural inductive bias over raw parameter count and provide reproducible guidance for multi-step financial forecasting.

Related papers

• VertiFuseX: Generalizable Financial Forecasting via Multi-Stream Temporal Fusion [0.0]
  We propose VertiFuseX, a hybrid LSTM architecture using penultimate-layer vertical fusion of multi-scale temporal representations.<n>We evaluate it on 15 years (2010-2024) of closing prices from 10 global equity indices using strict chronological out-of-sample testing.<n>VertiFuseX achieves 30-54% MAPE reductions and over 40% improvements in MAE and RMSE versus LSTM-based baselines.
  arXiv  Detail & Related papers  (2026-09-11T12:54:25Z)
• Systematic Evaluation of Learning Rate Scheduling Strategies Across Heterogeneous Architectures [48.83701310501069]
  We investigate the impact of schedulers on classification accuracy across a diverse pool of architectures.<n>We applied 25 scheduler configurations across nine PyTorch families, evaluating a total of 3,938 model variants on CIFAR-10.<n>The results show that the choice of scheduler depends heavily on the architecture: CosineAnnealingWarmRestarts and CyclicLR consistently outperform basic decay strategies.
  arXiv  Detail & Related papers  (2026-07-09T14:06:39Z)
• RAVEN: A Regime-Aware Variable-context Expert Network for Financial Time Series Forecasting [83.06074370551887]
  A fixed context window is mismatched to the time-varying optimal look-back of non-stationary price processes.<n>We propose a Mixture-of-Experts framework designed to adaptively determine the temporal context for each input sample.<n> Experiments on cumulative log-return prediction (HS300, S&P500) and fund sales forecasting demonstrate that RAVEN achieves SOTA performances, improves Pearson correlation by 9.2% on HS300 and 20.2% on S&P500, and reduces MSE by 18.2% on fund sales forecasting.
  arXiv  Detail & Related papers  (2026-06-23T02:11:04Z)
• EEVEE: Towards Test-time Prompt Learning in the Real World for Self-Improving Agents [64.96332056338923]
  EEVEE is the first multi-dataset test-time prompt learning framework for LLM agents.<n>It enables test-time prompt learning under real-world task streams.<n>EEVEE improves average multi-benchmark scores by 10.38 and 24.32 points over Qwen3-4B-Instruct and DeepSeek-V3.2.
  arXiv  Detail & Related papers  (2026-06-09T17:57:16Z)
• TEMPO: Scaling Test-time Training for Large Reasoning Models [87.61789183311856]
  Test-time training (TTT) adapts model parameters on unlabeled test instances during inference time.<n>We propose TEMPO, a TTT framework that interleaves policy refinement on unlabeled questions with periodic critic recalibration on a labeled dataset.
  arXiv  Detail & Related papers  (2026-04-21T10:01:04Z)
• MinerU2.5-Pro: Pushing the Limits of Data-Centric Document Parsing at Scale [92.09717763663873]
  We present MinerU2.5-Pro, which advances the state of the art purely through data engineering and training strategy design.<n>At its core is a Data Engine co-designed around coverage, informativeness, and annotation accuracy.<n>We show that MinerU2.5-Pro achieves 95.69 on OmniDocBench v1.6, improving over the same-architecture baseline by 2.71 points and surpassing all existing methods.
  arXiv  Detail & Related papers  (2026-04-06T15:44:18Z)
• Towards a Science of Scaling Agent Systems [79.64446272302287]
  We formalize a definition for agent evaluation and characterize scaling laws as the interplay between agent quantity, coordination structure, modelic, and task properties.<n>We derive a predictive model using coordination metrics, that cross-validated R2=0, enabling prediction on unseen task domains.<n>We identify three effects: (1) a tool-coordination trade-off: under fixed computational budgets, tool-heavy tasks suffer disproportionately from multi-agent overhead, and (2) a capability saturation: coordination yields diminishing or negative returns once single-agent baselines exceed 45%.
  arXiv  Detail & Related papers  (2025-12-09T06:52:21Z)
• Hybrid Quantum-Classical Ensemble Learning for S\&P 500 Directional Prediction [0.2538209532048867]
  We introduce a hybrid ensemble framework combining quantum sentiment analysis, Decision Transformer architecture, and strategic model selection.<n>We achieve 60.14% directional accuracy on S&P 500 prediction, a 3.10% improvement over individual models.
  arXiv  Detail & Related papers  (2025-12-06T22:22:09Z)
• Test time training enhances in-context learning of nonlinear functions [51.56484100374058]
  Test-time training (TTT) enhances model performance by explicitly updating designated parameters prior to each prediction.<n>We investigate the combination of TTT with in-context learning (ICL), where the model is given a few examples from the target distribution at inference time.
  arXiv  Detail & Related papers  (2025-09-30T03:56:44Z)
• FTSCommDetector: Discovering Behavioral Communities through Temporal Synchronization [15.690768429709811]
  Traditional community detection methods fail to capture synchronization-desynchronization patterns where entities move independently yet align during critical moments.<n>We introduce FTSCommDetector, implementing our Temporal Coherence Architecture (TCA) to discover similar and dissimilar communities.<n>As a result, FTSCommDetector achieves consistent improvements across four diverse financial markets.
  arXiv  Detail & Related papers  (2025-09-17T16:12:54Z)
• Comprehensive Modeling Approaches for Forecasting Bitcoin Transaction Fees: A Comparative Study [0.0]
  This study presents a systematic evaluation of six predictive models for forecasting Bitcoin transaction fees.<n>Our approach integrates comprehensive feature engineering spanning mempool metrics, network parameters, and historical fee patterns.<n>Traditional statistical approaches outperform more complex deep learning architectures.
  arXiv  Detail & Related papers  (2025-02-03T03:52:07Z)
• Accurate and Reliable Predictions with Mutual-Transport Ensemble [46.368395985214875]
  We propose a co-trained auxiliary model and adaptively regularizes the cross-entropy loss using Kullback-Leibler (KL) We show that MTE can simultaneously enhance both accuracy and uncertainty calibration. For example, on the CIFAR-100 dataset, our MTE method on ResNet34/50 achieved significant improvements compared to previous state-of-the-art method.
  arXiv  Detail & Related papers  (2024-05-30T03:15:59Z)
• LOT: Layer-wise Orthogonal Training on Improving $\ell_2$ Certified Robustness [14.206377940235091]
  Recent studies show that training deep neural networks (DNNs) with Lipschitz constraints are able to enhance adversarial robustness and other model properties such as stability. We propose a layer-wise orthogonal training method (LOT) to effectively train 1-Lipschitz convolution layers. We show that LOT significantly outperforms baselines regarding deterministic l2 certified robustness, and scales to deeper neural networks.
  arXiv  Detail & Related papers  (2022-10-20T22:31:26Z)
• SmoothNets: Optimizing CNN architecture design for differentially private deep learning [69.10072367807095]
  DPSGD requires clipping and noising of per-sample gradients. This introduces a reduction in model utility compared to non-private training. We distilled a new model architecture termed SmoothNet, which is characterised by increased robustness to the challenges of DP-SGD training.
  arXiv  Detail & Related papers  (2022-05-09T07:51:54Z)
• ERNIE-SPARSE: Learning Hierarchical Efficient Transformer Through Regularized Self-Attention [48.697458429460184]
  Two factors, information bottleneck sensitivity and inconsistency between different attention topologies, could affect the performance of the Sparse Transformer. This paper proposes a well-designed model named ERNIE-Sparse. It consists of two distinctive parts: (i) Hierarchical Sparse Transformer (HST) to sequentially unify local and global information, and (ii) Self-Attention Regularization (SAR) to minimize the distance for transformers with different attention topologies.
  arXiv  Detail & Related papers  (2022-03-23T08:47:01Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.