Using Tweets and Transaction Volume in Bitcoin Volatility Forecasts

Summary

This study models minute-level Bitcoin return volatility with generalized autoregressive conditional heteroskedasticity methods. It combines historical returns and volatility, as represented by variants of the GARCH family, with the Mixture of Distribution Hypothesis: market information flow can help explain return dynamics. Tweets related to Bitcoin and transaction volume serve as external information signals.

The paper tests whether these signals improve volatility forecasts and uses out-of-sample statistical tests to compare model variants. The simplest GARCH(1,1) model responds best to the addition of external signals in this evaluation. The available description does not report the sample dates, detailed performance measures, or the precise construction and timing of the tweet and volume variables. The result therefore supports the tested specification and data setup, but does not by itself show that these signals will improve forecasts in other periods or trading settings.

Key ideas

• The study forecasts Bitcoin volatility from minute-level price returns using GARCH-family models.
• It tests whether Bitcoin-related tweets and transaction volume add information about volatility.
• The approach frames return dynamics in terms of both historical volatility and market information flow.
• Out-of-sample tests find that GARCH(1,1) benefits most from adding the external signals among the models tested.
• The description gives limited detail about signal construction and the size of forecast improvements.

Tags

# Time-varying volatility in Bitcoin market and information flow at minute-level frequency


# Time-varying volatility in Bitcoin market and information flow at minute-level frequency









In this paper, we analyze the time-series of minute price returns on the Bitcoin market through the statistical models of generalized autoregressive conditional heteroskedasticity (GARCH) family. Several mathematical models have been proposed in finance, to model the dynamics of price returns, each of them introducing a different perspective on the problem, but none without shortcomings. We combine an approach that uses historical values of returns and their volatilities - GARCH family of models, with a so-called "Mixture of Distribution Hypothesis", which states that the dynamics of price returns are governed by the information flow about the market. Using time-series of Bitcoin-related tweets and volume of transactions as external information, we test for improvement in volatility prediction of several GARCH model variants on a minute level Bitcoin price time series. Statistical tests show that the simplest GARCH(1,1) reacts the best to the addition of external signal to model volatility process on out-of-sample data.

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.