Deep Reinforcement Learning for Cryptocurrency Trading: Practical Approach to Address Backtest Overfitting

• URL: http://arxiv.org/abs/2209.05559v2
• Date: Wed, 14 Sep 2022 08:11:32 GMT
• Title: Deep Reinforcement Learning for Cryptocurrency Trading: Practical Approach to Address Backtest Overfitting
• Authors: Berend Jelmer Dirk Gort, Xiao-Yang Liu, Xinghang Sun, Jiechao Gao, Shuaiyu Chen, Christina Dan Wang
• Abstract summary: We propose a practical approach to address backtest overfitting for cryptocurrency trading using deep reinforcement learning. We train the DRL agents, estimate the probability of overfitting, and reject the overfitted agents, increasing the chance of good trading performance. We show that the less overfitted deep reinforcement learning agents have a higher Sharpe ratio than that of more over-fitted agents, an equal weight strategy, and the S&P DBM Index.
• Score: 15.69458914236069
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Designing profitable and reliable trading strategies is challenging in the highly volatile cryptocurrency market. Existing works applied deep reinforcement learning methods and optimistically reported increased profits in backtesting, which may suffer from the false positive issue due to overfitting. In this paper, we propose a practical approach to address backtest overfitting for cryptocurrency trading using deep reinforcement learning. First, we formulate the detection of backtest overfitting as a hypothesis test. Then, we train the DRL agents, estimate the probability of overfitting, and reject the overfitted agents, increasing the chance of good trading performance. Finally, on 10 cryptocurrencies over a testing period from 05/01/2022 to 06/27/2022 (during which the crypto market crashed two times), we show that the less overfitted deep reinforcement learning agents have a higher Sharpe ratio than that of more over-fitted agents, an equal weight strategy, and the S&P DBM Index (market benchmark), offering confidence in possible deployment to a real market.

Related papers

• Predicting Bitcoin Market Trends with Enhanced Technical Indicator Integration and Classification Models [6.39158540499473]
  This study presents a machine learning model based on classification to forecast the direction of the cryptocurrency market. It is trained using historical data and important technical indicators such as the Moving Average Convergence Divergence, the Relative Strength Index, and Bollinger Bands. The results show a buy/sell signal accuracy of over 92%.
  arXiv  Detail & Related papers  (2024-10-09T14:29:50Z)
• Cryptocurrency Price Forecasting Using XGBoost Regressor and Technical Indicators [2.038893829552158]
  This study introduces a machine learning approach to predict cryptocurrency prices. We make use of important technical indicators such as Exponential Moving Average (EMA) and Moving Average Convergence Divergence (MACD) to train and feed the XGBoost regressor model. We evaluate the model's performance through various simulations, showing promising results.
  arXiv  Detail & Related papers  (2024-07-16T14:41:27Z)
• Beyond Trend Following: Deep Learning for Market Trend Prediction [49.89480853499917]
  We advocate for the use of Artificial Intelligence and Machine Learning techniques to predict future market trends. These predictions, when done properly, can improve the performance of asset managers by increasing returns and reducing drawdowns.
  arXiv  Detail & Related papers  (2024-06-10T11:42:30Z)
• Markowitz Meets Bellman: Knowledge-distilled Reinforcement Learning for Portfolio Management [7.199922073535674]
  This paper introduces a hybrid approach combining Markowitz's portfolio theory with reinforcement learning. In particular, our proposed method, called KDD (Knowledge Distillation DDPG), consist of two training stages: supervised and reinforcement learning stages. A comparative analysis against standard financial models and AI frameworks, using metrics like returns, the Sharpe ratio, and nine evaluation indices, reveals our model's superiority.
  arXiv  Detail & Related papers  (2024-05-08T22:54:04Z)
• Cryptocurrency Portfolio Optimization by Neural Networks [81.20955733184398]
  This paper proposes an effective algorithm based on neural networks to take advantage of these investment products. A deep neural network, which outputs the allocation weight of each asset at a time interval, is trained to maximize the Sharpe ratio. A novel loss term is proposed to regulate the network's bias towards a specific asset, thus enforcing the network to learn an allocation strategy that is close to a minimum variance strategy.
  arXiv  Detail & Related papers  (2023-10-02T12:33:28Z)
• Diffusion Variational Autoencoder for Tackling Stochasticity in Multi-Step Regression Stock Price Prediction [54.21695754082441]
  Multi-step stock price prediction over a long-term horizon is crucial for forecasting its volatility. Current solutions to multi-step stock price prediction are mostly designed for single-step, classification-based predictions. We combine a deep hierarchical variational-autoencoder (VAE) and diffusion probabilistic techniques to do seq2seq stock prediction. Our model is shown to outperform state-of-the-art solutions in terms of its prediction accuracy and variance.
  arXiv  Detail & Related papers  (2023-08-18T16:21:15Z)
• Profitable Strategy Design by Using Deep Reinforcement Learning for Trades on Cryptocurrency Markets [2.741266294612776]
  We have applied Proximal Policy Optimization, Soft Actor-C Imitation and Generative Adversarialritic Learning to strategy design problem of three cryptocurrency markets. Our test results on unseen data shows a great potential for this approach in helping investors with an expert system to exploit the market and gain profit.
  arXiv  Detail & Related papers  (2022-01-15T18:45:03Z)
• Deep Reinforcement Learning in Quantitative Algorithmic Trading: A Review [0.0]
  Deep Reinforcement Learning agents proved to be to a force to be reckon with in many games like Chess and Go. This paper reviews the progress made so far with deep reinforcement learning in the subdomain of AI in finance. We conclude that DRL in stock trading has showed huge applicability potential rivalling professional traders under strong assumptions.
  arXiv  Detail & Related papers  (2021-05-31T22:26:43Z)
• The Doge of Wall Street: Analysis and Detection of Pump and Dump Cryptocurrency Manipulations [50.521292491613224]
  This paper performs an in-depth analysis of two market manipulations organized by communities over the Internet: The pump and dump and the crowd pump. The pump and dump scheme is a fraud as old as the stock market. Now, it got new vitality in the loosely regulated market of cryptocurrencies. We report on three case studies related to pump and dump groups.
  arXiv  Detail & Related papers  (2021-05-03T10:20:47Z)
• Robust Deep Reinforcement Learning through Adversarial Loss [74.20501663956604]
  Recent studies have shown that deep reinforcement learning agents are vulnerable to small adversarial perturbations on the agent's inputs. We propose RADIAL-RL, a principled framework to train reinforcement learning agents with improved robustness against adversarial attacks.
  arXiv  Detail & Related papers  (2020-08-05T07:49:42Z)
• Pump and Dumps in the Bitcoin Era: Real Time Detection of Cryptocurrency Market Manipulations [50.521292491613224]
  We perform an in-depth analysis of pump and dump schemes organized by communities over the Internet. We observe how these communities are organized and how they carry out the fraud. We introduce an approach to detect the fraud in real time that outperforms the current state of the art.
  arXiv  Detail & Related papers  (2020-05-04T21:36:18Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.