Skip to content
All library documents

Using Cryptocurrency Exchange Flows to Forecast Intraday Returns and Volatility

Article arXiv papers · Author: Yeguang Chi et al.
Read the original

Summary

This study tests whether on-chain flows predict intraday returns and volatility for Bitcoin, Ethereum, and Tether. It examines one-to-six-hour intervals over 2017–2023 and distinguishes flows into exchanges from flows among investors’ wallets. The authors also use case studies to illustrate return forecasts and develop option strategies based on Ethereum net inflows to assess profit and loss on Ethereum investments.

The reported relationships vary by asset and outcome. Tether inflows to exchanges positively predict Bitcoin and Ethereum returns at multiple intervals, while Ethereum net inflows negatively predict Ethereum returns and volatility throughout the tested intervals. Bitcoin net inflows generally do not predict Bitcoin returns, except at one interval, but are negatively associated with volatility. The results are sample-specific predictive findings; the supplied summary does not describe the flow-data construction, option-strategy assumptions, trading costs, or whether the relationships persist outside the study period.

Key ideas

  • The analysis tests on-chain flow signals for intraday returns and volatility in Bitcoin, Ethereum, and Tether.
  • Tether inflows to exchanges are reported to predict higher Bitcoin and Ethereum returns at multiple intervals.
  • Ethereum net inflows are reported to predict lower Ethereum returns and volatility across the tested intervals.
  • Bitcoin net inflows generally lack return predictive power but are negatively associated with Bitcoin volatility.
  • The study examines Ethereum options strategies based on Ethereum net inflows.

Tags

CryptoOn-chain dataVolatilityOptionsStatistics
Full text
# Return and Volatility Forecasting Using On-Chain Flows in Cryptocurrency Markets


# Return and Volatility Forecasting Using On-Chain Flows in Cryptocurrency Markets









We empirically examine the intraday return- and volatility-forecasting power of on-chain flow data for Bitcoin(BTC), Ethereum(ETH), and Tether(USDT). We find ETH net inflows to strongly predict ETH returns and volatility in the 2017-2023 period. Our intraday frequencies are 1-6 hours. We find that differing significantly from forecasting patterns for BTC, ETH net inflows negatively predict ETH returns and volatility. First, we find that USDT flowing out of investors wallets and into cryptocurrency exchanges, namely, USDT net inflows into the exchanges, positively predicts BTC and ETH returns at multiple intervals and negatively predicts ETH volatility at various intervals and BTC volatility at the 6-hour interval. Second, we find that ETH net inflows negatively predict ETH returns and volatility for all intraday intervals. Third, BTC net inflows generally lack predictive power for BTC returns(except at 4 hours) but are negatively associated with volatility across all intraday intervals. We illustrate our findings on return forecasting via case studies. Moreover, we develop option strategies to assess profits and losses on ETH investments based on ETH net inflows. Our findings contribute to the growing literature on on-chain activity and its asset pricing implications, offering economically relevant insights for intraday portfolio management in cryptocurrency markets.

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.