Skip to content
All library documents

Classifying Cryptocurrency Market Regimes with a Convolutional Autoencoder

Article arXiv papers · Author: Vladimir Puzyrev
Read the original

Summary

The study uses a convolutional autoencoder to extract dominant patterns in cryptocurrency market behavior and group 40 cryptocurrencies into classes. It examines twelve consecutive six-month periods beginning on 15 May 2013, then relates changes in class membership over time to the maturation of the cryptocurrency market. The proposed use for traders is to interpret these evolving groups as a way to understand market structure and potentially inform investment or trading decisions.

The document provides a description of the method, asset universe, and observation windows, but no classification details, validation measures, specific regime results, or tested trading rules. It therefore offers an exploratory machine-learning approach to describing crypto market evolution rather than evidence of a profitable strategy. Class transitions are linked to market maturity, but the summary does not explain how that interpretation was established or whether the groupings remain stable under alternative data and model choices.

Key ideas

  • A convolutional autoencoder is used to extract dominant patterns from cryptocurrency markets.
  • The analysis classifies 40 cryptocurrencies into groups across twelve six-month periods.
  • Changes in group membership are interpreted in relation to market maturation.
  • The work suggests possible investment relevance but provides no direct trading test or reported validation metrics.

Tags

CryptoMachine learningStatistics
Full text
# Deep convolutional autoencoder for cryptocurrency market analysis


# Deep convolutional autoencoder for cryptocurrency market analysis









This study attempts to analyze patterns in cryptocurrency markets using a special type of deep neural networks, namely a convolutional autoencoder. The method extracts the dominant features of market behavior and classifies the 40 studied cryptocurrencies into several classes for twelve 6-month periods starting from 15th May 2013. Transitions from one class to another with time are related to the maturement of cryptocurrencies. In speculative cryptocurrency markets, these findings have potential implications for investment and trading strategies.

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.