Bitcoin during COVID-19 : Could Google Searches Explain Its Performance?

Purpose – This paper tries to contribute to financial modelling and decision-making in investment strategies by presenting that sentiment analysis using Google trends can explain the performance of Bitcoin (BTC) during the COVID-19 pandemic period (1/1/2020-26/3/2021). The information from social media regarding people’s interest could be beneficial for investors and practitioners. Design/methodology/approach – We employ a code which measures and indexes the Google searches for terms related to COVID-19 and BTC. These indices are included in a GARCH(1,1) model as explanatory variables in order to enable us to identify the variables that influenced the BTC performance during the COVID-19 period.Findings – The empirical findings show that COVID-19 fear increases risk aversion and has a negative impact on BTC prices. The increased interest regarding BTC, as expressed by Google searches, has a positive impact on the BTC prices. However, when this information arrives with a 1-day lag, the impact is negative because the increased interest of the previous day has already been incorporated in the previous day’s BTC price. Research limitations/implications – This study focuses on the number of Google searches. Further research on sentiment analysis should be done using textual analysis. This will allow for more accurate sentiment analysis and thus more accurate financial models which will include not only economic variables, but behavioral also. Practical implications – This note could be useful for scholars and practitioners who are interesting in financial modelling, investments, and decision-making. The methodological approach can be applied in any asset class, e.g. stocks, bonds, currencies etc., thus, there are no significant limitations to our research