Predicting crypto-currencies using sparse non-Gaussian state space models

TL;DR

This paper develops a Bayesian time-varying parameter VAR model with non-Gaussian features to forecast crypto-currency returns, demonstrating improved predictive performance over naive benchmarks and assessing economic gains through a trading simulation.

Contribution

It introduces a novel sparse non-Gaussian state space model with stochastic volatility for crypto return forecasting, incorporating Bayesian shrinkage for model flexibility.

Findings

Models outperform the random walk benchmark in real-time forecasts.

Forecasting gains translate into potential trading profits.

The approach captures key features of crypto return dynamics.

Abstract

In this paper we forecast daily returns of crypto-currencies using a wide variety of different econometric models. To capture salient features commonly observed in financial time series like rapid changes in the conditional variance, non-normality of the measurement errors and sharply increasing trends, we develop a time-varying parameter VAR with t-distributed measurement errors and stochastic volatility. To control for overparameterization, we rely on the Bayesian literature on shrinkage priors that enables us to shrink coefficients associated with irrelevant predictors and/or perform model specification in a flexible manner. Using around one year of daily data we perform a real-time forecasting exercise and investigate whether any of the proposed models is able to outperform the naive random walk benchmark. To assess the economic relevance of the forecasting gains produced by the proposed models we moreover run a simple trading exercise.