Comparison of different exit scenarios from the lock-down for COVID-19 epidemic in the UK and assessing uncertainty of the predictions

Anatoly Zhigljavsky; Roger Whitaker; Ivan Fesenko; Yakov Kremnitzer; Jack Noonan

In a mild case, the person stays infectious for τ M days and then discharges alive (that is, stops being infectious). We assume that the continuous version of τ M has Erlang distribution with shape parameter k M and rate parameter λ M , the mean of this distribution is k/λ M = 1/σ M (in simulations, we discretise the numbers to their nearest integers). We use values k M = 3 and λ M = 1/2 so that Eτ M = k M /λ M = 1/σ M = 6. The variance of τ M is var(τ M ) = k M /λ 2 M ; for k M = 3 and λ M = 1/2 we have var(τ M ) = 12 which seems to be a reasonable value.