Paper: arXiv 2602.00037

Authors: Yuanhong Wu, Wei Ye, Jingyan Xu, D. Frank Hsu

Abstract

In this work, we propose to apply a new model fusion and learning paradigm, known as Combinatorial Fusion Analysis (CFA), to the field of Bitcoin price prediction. Price prediction of financial product has always been a big topic in finance, as the successful prediction of the price can yield significant profit. Every machine learning model has its own strength and weakness, which hinders progress toward robustness. CFA has been used to enhance models by leveraging rank-score characteristic (RSC) function and cognitive diversity in the combination of a moderate set of diverse and relatively well-performed models. Our method utilizes both score and rank combinations as well as other weighted combination techniques. Key metrics such as RMSE and MAPE are used to evaluate our methodology performance. Our proposal presents a notable MAPE performance of 0.19%. The proposed method greatly improves upon individual model performance, as well as outperforms other Bitcoin price prediction models.

Complexity vs Empirical Score

  • Math Complexity: 4.5/10
  • Empirical Rigor: 4.0/10
  • Quadrant: Philosophers — conceptual discussion, limited math and data

Why this score: The paper applies Combinatorial Fusion Analysis (CFA) to ensemble standard ML models for daily Bitcoin price prediction, showing moderate novelty in method application. However, the empirical methodology suffers from basic quant flaws, such as using forward-filled future-overlapping features, relying on simple train/test splits without proper walk-forward validation, and evaluating level predictions via MAPE/RMSE rather than trading metrics.

Research Flowchart

  flowchart TD
    A["Research Goal: Predict Bitcoin Prices accurately using Model Fusion"] --> B["Data Inputs: Historical Bitcoin Price & Financial Indicators"]
    B --> C["Individual ML Models: Train multiple diverse predictive models"]
    C --> D["Combinatorial Fusion Analysis (CFA): Apply Score/Rank combinations & RSC Functions"]
    D --> E["Evaluation: Measure performance using RMSE and MAPE"]
    E --> F["Key Outcome: Outperforms individual & existing models (MAPE: 0.19%)"]

Cited References (37)

  1. Next-day bitcoin price forecast (2019)
  2. Bitcoin price prediction using machine learning: An approach to sample dimension engineering (2020)
  3. Analysis of bitcoin price prediction using machine learning (2023)
  4. Bitcoin price prediction: A machine learning sample dimension approach (2023)
  5. Machine learning models comparison for bitcoin price prediction (2018)
  6. A comparative study of bitcoin price prediction using deep learning (2019)
  7. A deep learning-based cryptocurrency price prediction scheme for financial institutions (2020)
  8. A novel cryptocurrency price prediction model using GRU , LSTM and bi- LSTM machine learning algorithms (2021)
and 29 more
  1. Forecasting the price of Bitcoin using deep learning (2021)
  2. Combinatorial fusion for improving portfolio performance (2010)
  3. Equity Ranking with Stochastic Dominance and Combinatorial Fusion (2023)
  4. Online Network DoS/DDoS Detection: Sampling, Change Point Detection, and Machine Learning Methods (2024)
  5. On the combination of two visual cognition systems using combinatorial fusion (2015)
  6. Long-term portfolio management using attribute selection and combinatorial fusion (2018)
  7. Combining multiple scoring systems for target tracking using rank–score characteristics (2009)
  8. Rank-score characteristics (RSC) function and cognitive diversity (2010)
  9. Bitcoin: A peer-to-peer electronic cash system (2008)
  10. Some simple economics of stablecoins (2022)
  11. A stacking ensemble deep learning model for bitcoin price prediction using Twitter comments on bitcoin (2022)
  12. Cryptocurrency Price Prediction Using Frequency Decomposition and Deep Learning (2023)
  13. The empirical analysis of bitcoin price prediction based on deep learning integration method (2022)
  14. A deep learning-based cryptocurrency price prediction model that uses on-chain data (2022)
  15. Enhancing ADMET property models performance through combinatorial fusion analysis (2023)
  16. Combinatorial Fusion Analysis (2024)
  17. Multi-layer combinatorial fusion using cognitive diversity (2021)
  18. Improving data and prediction quality of high-throughput perovskite synthesis with model fusion (2021)
  19. Backpropagation applied to handwritten zip code recognition (1989)
  20. Long Short-Term Memory (1997)
  21. A training algorithm for optimal margin classifiers (1992)
  22. Random Forests (2001)
  23. Methods of data fusion in information retrieval: rank vs. score combination (2002)
  24. Cognitive diversity: a measurement of dissimilarity between multiple scoring systems (2019)
  25. A deep learning-based LSTM for stock price prediction using twitter sentiment analysis (2024)
  26. Combinatorial fusion analysis: methods and practices of combining multiple scoring systems (2006)
  27. Enhancing ML-based DOS attack detection through combinatorial fusion analysis (2023)
  28. Xgboost: A scalable tree boosting system (2016)
  29. Analyzing the Effectiveness of Machine Learning Models for Next Day Cryptocurrency Price Prediction (2024)

0 of 37 cited works are in the corpus; resolved via arXiv source bundles, Semantic Scholar and OpenAlex. Corpus-wide: 2518 linked papers.