Modelling the effect of lockdown

抄録

We have modelled the effect of the lockdown during the first wave of COVID-19. We used SEIR type of model with a certain time lag between infection and becoming infectious. Firstly we confirmed the timing of the change of the coefficient of infection, growth rate of confirmed cases, corresponds to the change of mobility index, and secondly we assume the change of the coefficient of infection, activity index β (analogous to R₀) and fit the parameter to reproduce the actual number of confirmed cases. Finally, we assume that the activity index β is roughly proportional to the square of the mobility and fit the parameters. The curves in various countries fits reasonably well in any cases, but estimating β from various parameters (including temperature) in the long term remains as an important task.