Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Bitcoin Price Prediction using Machine Learning and Combinatorial Fusion Analysis

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

TL;DR

This paper tackles robust Bitcoin next-day price prediction by introducing Combinatorial Fusion Analysis (CFA), a model-fusion framework that leverages score, rank, and rank-score functions alongside cognitive diversity to combine five diverse base models (SVM, Random Forest, XGBoost, CNN, LSTM). It generates per-day prediction distributions from these models and uses CFA to form an optimal price prediction, evaluating performance with RMSE and MAPE. The results show CFA achieving a MAPE of $0.19\%$ on the test set, substantially outperforming individual models and several prior studies, with average score combination delivering the strongest accuracy. The work demonstrates the value of distribution-based ensemble fusion for financial time-series forecasting and discusses avenues for improvement, including sentiment analysis and multi-layer CFA, while noting limitations related to distribution construction and data leakage concerns.

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.

Bitcoin Price Prediction using Machine Learning and Combinatorial Fusion Analysis

TL;DR

This paper tackles robust Bitcoin next-day price prediction by introducing Combinatorial Fusion Analysis (CFA), a model-fusion framework that leverages score, rank, and rank-score functions alongside cognitive diversity to combine five diverse base models (SVM, Random Forest, XGBoost, CNN, LSTM). It generates per-day prediction distributions from these models and uses CFA to form an optimal price prediction, evaluating performance with RMSE and MAPE. The results show CFA achieving a MAPE of on the test set, substantially outperforming individual models and several prior studies, with average score combination delivering the strongest accuracy. The work demonstrates the value of distribution-based ensemble fusion for financial time-series forecasting and discusses avenues for improvement, including sentiment analysis and multi-layer CFA, while noting limitations related to distribution construction and data leakage concerns.

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.
Paper Structure (18 sections, 2 equations, 5 figures, 3 tables)

This paper contains 18 sections, 2 equations, 5 figures, 3 tables.

Table of Contents

  1. Introduction
  2. Machine Learning Models
  3. Combinatorial Fusion Analysis
  4. Methodology
  5. Data Preparation
  6. Base Models
  7. SVM
  8. Random Forest
  9. XGBoost
  10. CNN
  11. LSTM
  12. Method of Combination
  13. Score Function, Rank Function, and Rank-Score Function
  14. Cognitive Diversity
  15. Combination
  16. ...and 3 more sections

Figures (5)

  • Figure 1: Historical Bitcoin price and data partitioning
  • Figure 2: Phase I of methodology workflow
  • Figure 3: Phase II of methodology workflow
  • Figure 4: Phase III of methodology workflow
  • Figure 5: Phase IV of methodology workflow