Skip to content
All library documents

Machine Learning Strategies for Short-Term Cryptocurrency Price Prediction

Article arXiv papers · Author: Laura Alessandretti et al.
Read the original

Summary

This study tests whether cryptocurrency market inefficiencies can be used to produce abnormal trading profits. It evaluates simple algorithmic strategies supported by machine learning on daily observations covering 1,681 cryptocurrencies from November 2015 through April 2018. The reported comparison finds that these strategies outperform standard benchmarks, suggesting that relatively straightforward predictive mechanisms may help anticipate short-term market movements.

The abstract does not identify the specific algorithms, features, benchmarks, or transaction-cost assumptions, and it supplies no performance figures. The reported result is therefore limited to the study’s stated sample and period; it does not establish that the strategies would remain profitable in other market conditions or after trading costs. The work is useful as a historical test of machine learning based cryptocurrency trading, while further detail is needed to assess reproducibility and implementation.

Key ideas

  • The study tests machine learning assisted strategies for short-term cryptocurrency price prediction.
  • It uses daily data for 1,681 cryptocurrencies from November 2015 through April 2018.
  • The reported strategies outperform the study’s standard benchmarks.
  • The abstract does not specify algorithms, input features, or transaction-cost treatment.
  • The stated findings are confined to the historical sample described.

Tags

CryptoMachine learningBacktestingMomentum
Full text
# Anticipating cryptocurrency prices using machine learning


# Anticipating cryptocurrency prices using machine learning









Machine learning and AI-assisted trading have attracted growing interest for the past few years. Here, we use this approach to test the hypothesis that the inefficiency of the cryptocurrency market can be exploited to generate abnormal profits. We analyse daily data for $1,681$ cryptocurrencies for the period between Nov. 2015 and Apr. 2018. We show that simple trading strategies assisted by state-of-the-art machine learning algorithms outperform standard benchmarks. Our results show that nontrivial, but ultimately simple, algorithmic mechanisms can help anticipate the short-term evolution of the cryptocurrency market.

Shown in full with attribution under the source's licence. Licence: abstract CC0

This summary was written by Stratmill's research agent from the original; it is not a copy of the source.