Potential Issue in loglike_to_density

[sergejkaiser]

I created a data frame with the dates and counts of the confirmed covid 19 cases in belgium. When I estimated R with the get_R function is zero. When I further analysed the package code, the root cause seems to be the transformations in the loglike_to_density function. For this dataset log likelihood values are negative and after loglike_to_density function they become all equal to 0.

library(earlyR)
library(dplyr)
library(tidyr)
library(lubridate)
cases_be <- data.frame( I= c(1, 2,    11,   9,    10,   52,   59,   115,  54,   71,   54,   89,  
                                                                               167,  305,  258,  362,  321,  329,  451,  556,  870,  1080, 1105,
                                                                               646,  1028, 1329, 2453, 2023, 3612), dates = c("2020-02-04", "2020-03-01", "2020-03-02", "2020-03-03", "2020-03-04",
"2020-03-05", "2020-03-06", "2020-03-07", "2020-03-08", "2020-03-09",
"2020-03-10", "2020-03-11", "2020-03-12", "2020-03-13", "2020-03-14",
"2020-03-15", "2020-03-16", "2020-03-17", "2020-03-18", "2020-03-19",
"2020-03-20", "2020-03-21", "2020-03-22", "2020-03-23", "2020-03-24",
"2020-03-25", "2020-03-26", "2020-03-27", "2020-03-28"))
  cases_be %>%
  select(dates, I) %>% mutate (dates = lubridate::ymd(dates)) %>%
  uncount(I) %>%
  pull(dates) -> local_case_dates_be
  local_case_dates_be %>%
  incidence( .,last_date = "2020-03-28") -> local_cases_be_final
 
Res <- earlyR::get_R(x=local_cases_be_final, si_mean = 7.5, si_sd = 3.4,max_R = 8)

[aswansyahputra]

Is there any solution available for this issue?