Conventional statistical models and deep learning

The gap

Evidence profile

Stated in the cells research gap and cells future research sections of the source papers, classified as general, spanning 3 journals.

Supporting evidence — 4 representative gaps

• An Investigation of The Contribution of Attention-Based Hybrid Deep Learning Models to Prediction Errors in Cryptocurrency Markets (2026) · Zenodo (CERN European Organization for Nuclear Research) · doi

  The study identifies a methodological gap in the literature, where multiple deep learning models are not benchmarked under identical datasets and experimental conditions. - The study addresses the lack of comparative analysis of deep learning architectures for predicting cryptocurrency prices.

  generalstated in cells research gapevidence 5/5
  Keywords: study identifies methodological gap literature multiple deep learning
• Deep learning-based short-term cryptocurrency price forecasting using LSTM models enhanced with technical indicators (2026) · International Journal of Pure and Applied Sciences · doi

  Traditional models have limited representational capabilities and their predictive performance degrades in the presence of structural breaks and long-term dependencies. - There is a need for a deep learning-based approach for short-term cryptocurrency price forecasting.

  generalstated in cells research gapevidence 5/5
  Keywords: traditional models have limited representational capabilities predictive performance
• Hybrid TCN-transformer Model with Multi-head Attention for Stock Price Forecasting (2026) · International Journal of Intelligent Systems and Applications · doi

  Conventional statistical models and deep learning techniques have limitations in capturing intricate relationships and patterns in stock market data. - The existing models are not able to handle the complexity and volatility of financial time series data effectively.

  generalstated in cells research gapevidence 5/5
  Keywords: conventional statistical models deep learning techniques have limitations
• A SYSTEMATIC LITERATURE REVIEW AND BIBLIOMETRIC ANALYSIS OF MACHINE LEARNING ALGORITHMS AND TECHNICAL INDICATORS TO STOCK PRICE PREDICTION (2026) · Zenodo (CERN European Organization for Nuclear Research) · doi

  Further research is needed to address the loopholes in model robustness, interpretability, and standard benchmarking. - The use of multi-source data fusion in financial forecasting should be explored. - The development of more effective stock price prediction models using deep learning and hybrid approaches should be investigated.

  generalstated in cells future researchevidence 4/5
  Keywords: further research needed address loopholes model robustness interpretability

Questions about this gap