Deep Temporal Convolutional Networks for High-Frequency Cryptocurrency Price Forecasting
Bibliographic record
Abstract
High-frequency cryptocurrency price prediction remains challenging due to extreme volatility, market noise, and complex temporal dependencies. Traditional machine learning methods, recurrent neural networks (RNNs), and attention-enhanced models have shown limited predictive accuracy and generalization on such data. To address these limitations, this study systematically evaluates Gated Recurrent Units (GRU), Long Short-Term Memory (LSTM), Bidirectional LSTM (BiLSTM), and Temporal Convolutional Networks (TCNs) using multi-factor inputs. The dataset spans 2018–2024 and includes OHLC prices, trading volume, number of trades, short-term buying pressure, and technical indicators such as moving averages, EMA deviations, MACD, Bollinger Bands, and momentum. Models were trained using standard procedures and evaluated via RMSE, MAE, MAPE, and directional accuracy. Results demonstrate that TCNs consistently outperform RNN-based models in both predictive precision and directional reliability. Multi-factor analysis reveals that price information dominates predictive performance, while volume, trade counts, and buying pressure provide complementary signals. These findings indicate that TCNs, leveraging dilated causal convolutions and residual connections, can effectively capture long-range temporal dependencies in high-frequency financial data. The study highlights the practical benefits of combining TCNs with rich multi-factor representations for robust and reliable cryptocurrency price forecasting.
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 | 0.002 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.001 | 0.001 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.001 | 0.001 |
| Open science | 0.001 | 0.000 |
| Research integrity | 0.000 | 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".