CRYPTOCURRENCY RETURNS’ CAUSALITY AND PREDICTION BY USING VECTOR ERROR CORECTION MODEL
CRYPTOCURRENCY RETURNS’ CAUSALITY AND PREDICTION BY USING VECTOR ERROR CORECTION MODEL

Summary/Abstract: The main goal of the research is to predict the future monthly returns of cryptocurrencies using the Vector Error Correction Model (VECM). Time series for the period 2018-2021 consists of data on monthly returns for the cryptocurrencies Bitcoin, Ethereum and Ripple, as well as monthly returns on gold and theS&P500 stock index. Within the VECM, using the Johansen and Granger tests, short-term cointegration and causality among variables were determined, without the existence of long-term equilibrium. The resulting model for short-term prediction of the monthly returns of the cryptocurrency Bitcoin was evaluated as unbiased and stable with a realistic forecast error of 0.168 (16.8%)

• Details
• Contents