Cryptoasset Return Factors and Cross-Sectional Statistical Arbitrage

Summary

This work studies factor models for the cross-section of daily returns across cryptocurrencies and other digital assets with exchange market data. It identifies a leading factor that appears to make a strong contribution to cryptoasset returns. The document also describes supporting materials for downloading data, computing risk factors, and conducting out-of-sample backtests.

The authors suggest that cross-sectional statistical arbitrage may be feasible in cryptoassets, conditional on efficient execution and the ability to short. The summary does not name the factor, describe its construction, report backtest metrics, or establish whether the opportunity survives costs and market frictions. Its practical implication is therefore a research lead rather than evidence of a deployable strategy.

Key ideas

• The study models daily returns across a broad set of exchange-traded cryptoassets.
• It identifies a leading factor associated with cross-sectional cryptoasset returns.
• The research includes out-of-sample backtesting as part of its empirical approach.
• Potential statistical arbitrage depends on efficient execution and access to shorting.

Tags

# Cryptoasset Factor Models


# Cryptoasset Factor Models









We propose factor models for the cross-section of daily cryptoasset returns and provide source code for data downloads, computing risk factors and backtesting them out-of-sample. In "cryptoassets" we include all cryptocurrencies and a host of various other digital assets (coins and tokens) for which exchange market data is available. Based on our empirical analysis, we identify the leading factor that appears to strongly contribute into daily cryptoasset returns. Our results suggest that cross-sectional statistical arbitrage trading may be possible for cryptoassets subject to efficient executions and shorting.

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.