Richardson-Lucy Deconvolution.
deconvolution_RL(
observed,
times,
p_delay,
max_iter = 50,
out_col_name = "RL_result",
right_censor = TRUE,
verbose = FALSE
)A data frame with columns time and out_col_name (which gives the imputed number of infections per timestep)
A vector of the number of observed cases, deaths, hospitalizations, etc. per day
A numeric vector of times corresponding to the entries in observed. Must be the same length as observed.
A numeric vector whose entries give the probability that the delay from infection to observation is exactly 0, 1, ... N days. p_delay must sum to 1 and to facilitate vectorization should be the same length as observed, and times (the last several entries will most likely be 0s).
Maximum number of times to iterate. Following Goldstein et al., the algorithm continues to run until the normalized chi2 statistic comparing the observed and expected number of deaths per day falls below 1, or until the maximum number of iterations is reached.
String giving the name of the column in which to output the imputed times of infection
If TRUE (default), upscale the observations
at the end of the time series based on the cumulative probability
that an infection occurring on that date would have been observed.
Logical. Verbose mode.