Modeling COVID-19 Infected Cases and Deaths Based on Generalized Method of Moments

Abstract

Abstract This paper investigates the dynamic relationships between the number of COVID-19 infected cases and deaths in all the districts of Karnataka state, India, from July 2020 to December 2021 based on the panel Generalized Method of Moments (GMM). The panel GMM model with the first difference transformation was found suitable for studying the dynamics of the number of deaths due to COVID-19 infections over time. The one-period lag (DEATHS (-1)) has a positive and significant effect on the number of deaths (DEATH). The Wald test confirms the validity of the coefficients' significance and adds explanatory power to the model. The correlation between number of fatalities at time t positively correlated with the number of deaths in the previous period. Also, the number of infected cases positively and significantly influences the number of deaths over time. Granger pairwise causality test reveals the existence of bi-directional causality relationships between the COVID-19 infected cases and deaths. JEL classification numbers: E18, HO, I1, J64, J88. Keywords: Arellano-Bond serial correlation test, Cross-sectional dependence test, Cointegration test, Granger causality test, Generalized Method of Moments, Kao cointegration test, Hausman test, Levin-Lin-Chu unit root test, Wald test.