Table 2 Prior distributions for the Bayesian statistical model.
From: Model-based evaluation of school- and non-school-related measures to control the COVID-19 pandemic
Parameter | Priora | Explanation |
|---|---|---|
ϵ | Uniform (0, 1) | Flat prior |
α | InvGamma (32.25, 9.75) | 99% of the prior density of 1/α between 2 and 5 days |
γ | InvGamma (22.6, 2.44) | 99% of the prior density of 1/γ between 5 and 15 days |
ν[0, 20), ν[20, 30) |  |  |
ν[30, 40), ν[40, 50) | folded-\({\mathcal{N}}(0,5)\) | Vague prior |
ν[50, 60), ν[60, 70) |  |  |
ν[70, 80), ν80+ |  |  |
β[0, 20)b | Lognormal (−1.47, 0.1) | Log-odds \(-1.47=\mathrm{log}\,(0.23)\) based on prior estimates8 |
β[20, 60)b | Lognormal (−0.45, 0.1) | Log-odds \(-0.45=\mathrm{log}\,(0.64)\) based on prior estimates8 |
r | Lognormal(5, 2) | Vague prior |
ζ1 | \({\mathcal{N}}(1,0.1)\) | A priori, we expect the reduction in contacts after the first lockdown to account for most of the decrease in the transmission rate |
t1 | \({\mathcal{N}}(23,7)\) | The mean of t1 is given by the day of initiation of most drastic social distancing measures (15 March); most measures were taken within two weeks from this date |
K1 | \({\rm{Exp}}(1)\) | For K1 = 1 the uptake of measures takes approximately 6 days |
θ | Uniform (10−7, 5 × 10−4) | Vague prior allowing for approximately 100–105 infections at time t0 |
