Nonlinear causality analysis of Bitcoin returns and COVID-19 pandemic

Sami, Janesh (2026) Nonlinear causality analysis of Bitcoin returns and COVID-19 pandemic. Computational Economics, 67 . pp. 485-509. ISSN 0927-7099

Text - Published Version Restricted to Repository staff only Download (1MB)

Abstract

The recent COVID-19 pandemic and ongoing geopolitical tensions have disrupted the functioning of the global financial system and heightened investment uncertainty. These developments have made risk mitigation, portfolio management, and regulatory responses increasingly challenging for investors, portfolio managers, and policymakers, respectively. Against this background, this paper empirically investigates the causal linkage between COVID-19 cases and Bitcoin returns using daily data from 12 March 2020 to 15 June 2021 by employing both linear Granger causality and a non-parametric Granger causality test. The empirical results reveal that there is no evidence of either linear or nonlinear Granger causality from total COVID-19 cases – whether in India or the United States, or globally – to Bitcoin returns. However, country-level analysis indicates the presence of a nonlinear Granger causal relationship from COVID-19 cases in China (but not India and the United States) to Bitcoin returns. Our findings indicates that even though the total number of COVID-19 cases in China has been well below that of the United States and India, it still has predictive content for Bitcoin returns. This finding is critical for ensuring a nuanced understanding of the Bitcoin market and underscores the possible roles of media narratives, speculations about the source of the COVID-19 pandemic, and market perceptions in the causal linkage between the pandemic and Bitcoin returns.

Item Type:	Journal Article
Subjects:	H Social Sciences > HG Finance
Divisions:	School of Accounting, Finance and Economics (SAFE)
Depositing User:	Ms Shalni Sanjana
Date Deposited:	29 Apr 2026 22:18
Last Modified:	29 Apr 2026 22:18
URI:	https://repository.usp.ac.fj/id/eprint/15357

Actions (login required)

View Item