Paper: arXiv 2407.18334
Authors: Abdul Jabbar, Syed Qaisar Jalil
Abstract
This study evaluates the performance of 41 machine learning models, including 21 classifiers and 20 regressors, in predicting Bitcoin prices for algorithmic trading. By examining these models under various market conditions, we highlight their accuracy, robustness, and adaptability to the volatile cryptocurrency market. Our comprehensive analysis reveals the strengths and limitations of each model, providing critical insights for developing effective trading strategies. We employ both machine learning metrics (e.g., Mean Absolute Error, Root Mean Squared Error) and trading metrics (e.g., Profit and Loss percentage, Sharpe Ratio) to assess model performance. Our evaluation includes backtesting on historical data, forward testing on recent unseen data, and real-world trading scenarios, ensuring the robustness and practical applicability of our models. Key findings demonstrate that certain models, such as Random Forest and Stochastic Gradient Descent, outperform others in terms of profit and risk management. These insights offer valuable guidance for traders and researchers aiming to leverage machine learning for cryptocurrency trading.
Complexity vs Empirical Score
- Math Complexity: 3.0/10
- Empirical Rigor: 6.0/10 (capped from 9.0: transaction costs not modelled; reported Sharpe ratio of 8.68 is implausibly high for daily or lower-frequency data; see penalty flags)
- Quadrant: Street Traders — empirical and applied, lower math complexity (effective empirical rigor capped from 9.0)
Why this score: The paper focuses on evaluating 41 standard ML models using accessible metrics like MAE and Sharpe Ratio, resulting in low math complexity, while its emphasis on backtesting, forward testing, and real-world trading scenarios ensures high empirical rigor.
Research Flowchart
flowchart TD
A["Research Goal<br>Evaluate ML models for Bitcoin trading"] --> B["Data Collection<br>Historical & recent Bitcoin price data"]
B --> C["Model Training<br>21 Classifiers & 20 Regressors<br>Random Forest, SGD, etc."]
C --> D["Evaluation Process<br>Backtesting, Forward Testing, Real-world Trading"]
D --> E{"Performance Metrics"}
E --> F["Machine Learning Metrics<br>MAE, RMSE"]
E --> G["Trading Metrics<br>PnL%, Sharpe Ratio"]
F & G --> H["Key Findings & Outcomes"]
H --> I["Random Forest & SGD<br>best for profit & risk"]
H --> J["Comprehensive guide for<br>effective trading strategies"]
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 evaluates 41 machine learning models for Bitcoin price prediction in algorithmic trading, assessing both predictive and trading performance.
Source sentence
This study evaluates the performance of 41 machine learning models, including 21 classifiers and 20 regressors, in predicting Bitcoin prices for algorithmic trading.
Assumes 3
- Market volatility affects trading outcomes. market
Source sentence
This includes considerations of market volatility, transaction costs, and other factors that impact on the trading outcomes.
- Transaction costs are relevant to model evaluation. execution
Source sentence
This includes considerations of market volatility, transaction costs, and other factors that impact on the trading outcomes.
- Historical Bitcoin price data is sufficient for training robust models. data
Source sentence
our study leverages a detailed dataset of Bitcoin prices, publicly available since the inception of Bitcoin trading in 2013.
Data
BTC/USDT · crypto · daily
- Bitcoin price dataset market-prices public
Source sentence
The training dataset spans a decade, from January 2013 to January 2023
Twitter / X
Costs
Unclear - sources disagree
Source sentence (the authors' words)
This includes considerations of market volatility, transaction costs, and other factors that impact on the trading outcomes.
Finds 3
Sharpe 8.68 · total return 15.4%
- Random Forest and Stochastic Gradient Descent models outperform others in profit and risk management. performance
Source sentence
Key findings demonstrate that certain models, such as Random Forest and Stochastic Gradient Descent, outperform others in terms of profit and risk management.
- Many models that performed well in backtesting did not translate effectively to forward tests and real-world scenarios. negative
Source sentence
Many models that performed well in backtesting did not translate effectively to forward tests and real-world scenarios, highlighting the limitations of relying solely on backtesting.
- Integrating both machine learning metrics and trading metrics provides a comprehensive assessment of model performance. performance
Source sentence
The integration of both machine learning metrics (e.g., Mean Absolute Error, Root Mean Squared Error) and trading metrics (e.g., Profit and Loss percentage, Sharpe Ratio) provided a comprehensive assessment of model performance.
Checked by others
No other paper in the archive has been found to support or contradict these findings (compared with 19 related papers). Coverage is partial, so this is neither replication nor refutation.
1 paper in the archive
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 (37)
- The Sharpe Ratio (1994)
- Mutual Fund Performance (1966)
- Optuna: A Next-generation Hyperparameter Optimization Framework (2019)
- Forecasting the movements of Bitcoin prices: an application of machine learning algorithms — → 2303.04642
- Empirical evaluation of machine learning performance in forecasting cryptocurrencies (2023)
- Analyzing the machine Learning methods to predict Bitcoin Pricing (2024)
- Height-end Multi-layer Perceptron and Machine Learning Methods of Forecasting Bitcoin Price Time Series (2023)
- Performance of machine learning methods in predicting trend in price and trading volume of cryptocurrencies (2023)
and 29 more
- Artificial intelligence techniques in financial trading: A systematic literature review (2024)
- Cryptocurrency trading using machine learning (2020)
- Forecasting and trading cryptocurrencies with machine learning under changing market conditions (2021)
- Machine learning for cryptocurrency market prediction and trading (2022)
- Real-world model for bitcoin price prediction (2022)
- Bitcoin USD Closing Price (BTC-USD) Comparison Using Simple Moving Average And Radial Basis Function Neural Network Methods (2022)
- Bitcoin price forecasting and trading: Data analytics approaches (2022)
- Cryptocurrency trading: a comprehensive survey (2022)
- Model-free reinforcement learning for financial portfolios: a brief survey (2019) — link
- Comparative performance of machine learning algorithms for cryptocurrency forecasting (2018)
- A comparative study for Bitcoin cryptocurrency forecasting in period 2017-2019 (2020)
- Time-series prediction of cryptocurrency market using machine learning techniques (2021)
- Your Sentiment Matters: A Machine Learning Approach for Predicting Regime Changes in the Cryptocurrency Market (2023)
- Predicting Bitcoin Prices Using Machine Learning (2023)
- Short-term bitcoin market prediction via machine learning (2021)
- Time-series forecasting of Bitcoin prices using high-dimensional features: a machine learning approach (2020)
- Bitcoin Price Prediction Using Machine Learning Algorithms (2023)
- Signal Prediction in Cryptocurrency Tradeoperations: A Machine Learning-Based Approach
- Forecasting mid-price movement of Bitcoin futures using machine learning (2021)
- Cryptocurrency Price Prediction Using Deep Learning and Machine Learning (2022)
- On Forecasting Cryptocurrency Prices: A Comparison of Machine Learning, Deep Learning, and Ensembles (2023)
- Daily cryptocurrency returns forecasting and trading via machine Learning (2021)
- Price movement prediction of cryptocurrencies using sentiment analysis and machine learning (2019)
- Bitcoin price prediction based on other cryptocurrencies using machine learning and time series analysis (2020)
- Comparative Analysis Of Machine Learning Models For Predicting Bitcoin Price Rate (2021)
- Analysis of bitcoin price prediction using machine learning (2023)
- Prediction of Bitcoin Price Movements Based on Machine Learning Method and Strategy Construction
- Are Bitcoins price predictable? Evidence from machine learning techniques using technical indicators (2019) — link
- Profitable Strategy Design for Trades on Cryptocurrency Markets with Machine Learning Techniques (2021) — link
1 of 37 cited works are in the corpus; resolved via arXiv source bundles, Semantic Scholar and OpenAlex. Corpus-wide: 2596 linked papers.