Abstract
Abstract New flexible-form and semi-parametric autoregressive non-linear count models for panel data are developed to analyse the spread and containment of the COVID-19 pandemic. The models are based on a discrete time form of the SIR model. These methods lead naturally to estimators of the infection process and daily reproduction numbers by jurisdiction. Two semi-parametric versions of the reproduction numbers are developed corresponding to currently popular parametric estimators. The estimators are applied to a large international data set to estimate these parameters for 221 jurisdictions at both national and subnational levels. JEL classification numbers: C14, C23, I18. Keywords: COVID-19, Reproduction Numbers, Panel Data, Count Models, Semi-parametric Approach.
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have