Paper: arXiv 2405.08089

Authors: Ali Mohammadjafari

Abstract

Prediction of stock prices has been a crucial and challenging task, especially in the case of highly volatile digital currencies such as Bitcoin. This research examineS the potential of using neural network models, namely LSTMs and GRUs, to forecast Bitcoin’s price movements. We employ five-fold cross-validation to enhance generalization and utilize L2 regularization to reduce overfitting and noise. Our study demonstrates that the GRUs models offer better accuracy than LSTMs model for predicting Bitcoin’s price. Specifically, the GRU model has an MSE of 4.67, while the LSTM model has an MSE of 6.25 when compared to the actual prices in the test set data. This finding indicates that GRU models are better equipped to process sequential data with long-term dependencies, a characteristic of financial time series data such as Bitcoin prices. In summary, our results provide valuable insights into the potential of neural network models for accurate Bitcoin price prediction and emphasize the importance of employing appropriate regularization techniques to enhance model performance.

Complexity vs Empirical Score

  • Math Complexity: 3.0/10
  • Empirical Rigor: 6.0/10
  • Quadrant: Street Traders — empirical and applied, lower math complexity

Why this score: The paper uses standard deep learning architectures (LSTMs/GRUs) with basic regularization (L2), representing accessible math, but employs concrete backtesting methodology (5-fold cross-validation) and reports specific performance metrics (MSE) on real financial data.

Research Flowchart

  flowchart TD
  A["Research Goal<br>Predict Bitcoin Price"] --> B{"Data Input"}
  B --> C["Time Series Data"]
  
  C --> D["Methodology<br>5-Fold Cross-Validation + L2 Regularization"]
  
  D --> E["Computational Process<br>Train LSTM & GRU Models"]
  
  E --> F["Outcome"]
  
  F --> G["GRU Model<br>MSE: 4.67"]
  F --> H["LSTM Model<br>MSE: 6.25"]
  
  G & H --> I["Key Finding<br>GRU outperforms LSTM for Bitcoin prediction"]

Evidence Card

What the authors claim, on what data, and who has checked it. Read from the paper's own text by a language model: every quote is verbatim, nothing was re-run. How to read this card

Proposes

The paper develops and compares two neural network models, LSTM and GRU, for Bitcoin price prediction.

Source sentence
We develop two different neural network models (LSTM and GRU) to predict Bitcoin prices and compare their performance.

Assumes 2

  • Five‑fold cross‑validation provides accurate and reliable performance estimates on unseen data. statistical
    Source sentence
    By using 5-fold cross-validation, we can obtain more accurate and reliable estimates of the model’s performance on unseen data
  • L2 regularization improves model generalization and reduces overfitting. model
    Source sentence
    we apply L2 regularization to the weights of the models.

Data

  • Bitcoin Historical Data dataset · Yahoo Finance market-prices public
    December 31, 2015 to April 6, 2023
    Source sentence
    The Bitcoin Historical Data dataset will be downloaded from Yahoo Finance website

Recognised sources: Yahoo Finance

Costs

Not applicable The paper reports no trading or execution result.

Finds 2

  • GRU model achieves lower mean squared error (MSE = 4.67) than LSTM model (MSE = 6.25) on the test set. predictive-accuracy
    Source sentence
    Specifically, the GRU model has an MSE of 4.67, while the LSTM model has an MSE of 6.25 when compared to the actual prices in the test set data.
  • GRU models offer better accuracy than LSTM models for predicting Bitcoin prices. performance
    Source sentence
    Our study demonstrates that the GRUs models offer better accuracy than LSTMs model for predicting Bitcoin's price.

Checked by others

No other paper in the archive has been found to support or contradict these findings (compared with 33 related papers). Coverage is partial, so this is neither replication nor refutation.

Code

No repository by the authors found in the paper.

Extracted 2026-11-14 · same card as JSON · A language model chose these items: check the source sentence before relying on one.

Cited References (11)

  1. Bitcoin Blockchain Explained: Development and Challenges (2022)
  2. Can we stabilize the price of a cryptocurrency? Understanding the design of Bitcoin and its potential to compete with Central Bank money (2022)
  3. Efficient binary grasshopper optimization based neural network algorithm for bitcoin value prediction (2022)
  4. Past, present, and future of the application of machine learning in cryptocurrency research (2022)
  5. Cryptocurrencies as a financial asset: A systematic analysis (2019)
  6. China’s commercial bank stock price prediction using a novel K-means-LSTM hybrid approach (2022)
  7. Financial time series forecasting with multi-modality graph neural network (2022)
  8. S_I_LSTM: stock price prediction based on multiple data sources and sentiment analysis (2022)
and 3 more
  1. Analysis of different artificial neural networks for Bitcoin price prediction (2022)
  2. A stacking ensemble deep learning model for bitcoin price prediction using Twitter comments on bitcoin (2022)
  3. Cryptocurrency price prediction with neural networks of LSTM and Bayesian optimization (2022)

0 of 11 cited works are in the corpus; resolved via arXiv source bundles, Semantic Scholar and OpenAlex. Corpus-wide: 2596 linked papers.