Hybrid Machine Learning Models for Time Series Forecasting: Integrating LSTM-Inspired Neural Residuals with Traditional Predictive Techniques in the Finance Domain
Bibliographic record
Abstract
Forecasting financial time series remains one of the most challenging problems in computational finance due to volatility, non-stationarity, and regime shifts. Cryptocurrencies, particularly Bitcoin, represent a unique testbed owing to their extreme fluctuations and structural breaks. In this paper, we propose a novel Hybrid Dynamic Error-Correcting Ensemble (DECE) framework that integrates classical econometric methods (ARIMA, Seasonal-Naïve) with neural residual learners inspired by long short-term memory (LSTM) architectures. Unlike conventional hybrids with static weights, DECE dynamically adjusts model contributions based on rolling error statistics, thereby adapting to shifts in volatility and market regimes. Using Bitcoin/USD data at both daily and 15-minute frequencies (2012–2025), we conduct rigorous evaluations across multiple error metrics (MAE, RMSE, MAPE, sMAPE) and statistical significance testing (Diebold– Mariano). Results indicate that while naïve benchmarks remain highly competitive at daily horizons, the proposed DECE hybrid significantly outperforms all baselines, particularly in intraday forecasting where volatility clustering and nonlinear effects dominate. Our findings highlight the importance of adaptive hybridization in financial forecasting, offering both theoretical insights and practical tools for traders, risk managers, and policymakers in highly dynamic financial environments.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.001 | 0.004 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.000 |
| Bibliometrics | 0.000 | 0.001 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.001 | 0.001 |
| Open science | 0.001 | 0.001 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.001 | 0.000 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".