Table 1 Parameter values used for method comparison
From: estimateR: an R package to estimate and monitor the effective reproductive number
R package | Parameter | Value | Notes |
|---|---|---|---|
estimateR | Smoothing parameter \(\sigma\) | 9 Time steps | |
Incubation period | Gamma (shape=3.2, scale = 2.1) | As specified in simulations | |
Observation delay | Gamma (shape=2.7, scale=2.6) | As specified in simulations | |
(from symptoms to confirmation) | |||
Generation time | Gamma (mean = 4.8, SD = 2.3) | As specified in simulations | |
Other parameters | Default settings | ||
epidemia | Generation time | Gamma (mean = 4.8, SD = 2.3) | As specified in simulations |
Observation model family | Negative binomial | Analyses failed with log-normal | |
(log-normal fits | |||
simulated noise) | |||
Delay distribution | Discretized convolution of | As specified in simulations | |
(from infections to observations) | Gamma (shape=3.2, scale = 2.1) | ||
and | |||
Gamma (shape=2.7, scale=2.6) | |||
Hyperprior scale | 0.2 | ||
on \(R_t\) random walk | |||
Other parameters | Default settings | ||
EpiNow2 | Incubation period | Log-normal(\(\mu\) = 1.68, \(\sigma\) = 0.63) | Log-normal fit of |
Gamma (shape=3.2, scale = 2.1) | |||
(used in simulations) | |||
Observation delay | |||
(from symptoms to confirmation) | Log-normal(\(\mu\) = 1.68, \(\sigma\) = 0.67) | Log-normal fit of | |
Gamma (shape=2.7, scale = 2.6) | |||
(used in simulations) | |||
Generation time | Gamma(mean = 4.8, SD = 2.3) | As specified in simulations | |
Gaussian process | Applied to global mean | ||
Observation model family | Negative binomial | ||
Other parameters | Default settings |