get_R: weird likelihood estimation of R

[jeskarp]

If I try to estimate the likelihood of R based on incidence data that has no cases at the end of the observation period, the maximum likelihood of R is estimated to be 0.

Below is an example using a measles outbreak from 1861.

You'll need these packages

library(outbreaks)
library(incidence)
library(earlyR)

We are using this dataset:
measles_hagelloch_1861

We then make an incidence object out of the measles data and take a subset that has few cases at the end of the observation period:

measles_hagelloch_incidence <- incidence(measles_hagelloch_1861$date_of_prodrome, interval = 1)
measles_hagelloch_incidence_sub <- measles_hagelloch_incidence[31:42, ]

And then attempt to estimate R for this observation window:
R_subset <- get_R(measles_hagelloch_incidence_sub, si_mean = 10.89, si_sd = 1.63, max_R = 10)

R is estimated to be 0 as seen in:
plot(R3)

[thibautjombart]

closed via ae5450a