Forecasting Bitcoin Futures: A Lasso-BMA Two-Step Predictor Selection for Investment and Hedging Strategies

【Author】 Huang, Weige; Gao, Xiang

【Source】SAGE OPEN

【影响因子】2.032

【Abstract】After Bitcoin futures were introduced by the Chicago Mercantile Exchange in December 2017, their trading volume has stayed in an uptrend due to speculation, though the scale is still small compared to other traditional futures. As increasing trading indicates more attention and the presence of institutional traders, there exists a need for reliable return and variance forecasts of Bitcoin futures contracts. Therefore, this paper first applies LASSO to pick out best-fitting predictors by shrinking the dimension of a universe of potential determinants sourced from intraday Bitcoin spot trades and daily futures variables. Then, a second round of predictor selection is conducted via Bayesian model averaging so that the modeling uncertainty can be mitigated. We find that factors standing out from this two-step procedure possess a strong predictive power for Bitcoin futures return and volatility in different time horizons. It is further demonstrated that the investment and hedging strategies established based on our forecasts perform well in out-of-sample validations.

【Keywords】Bitcoin; cryptocurrency; futures; LASSO; Bayesian model averaging; return prediction; volatility prediction

【发表时间】2023 JAN

【收录时间】2023-09-19

【文献类型】实证数据

【主题类别】

区块链治理-市场治理-市场分析