R/utils.R

In rmorsomme/PDSIR: Data-augmentation Marko chain Monte Carlo for Fitting the Stochastic SIR Model to Incidence Counts

utils::globalVariables(c("."))

#' Include beta to the parameter vector theta
#'
#' @param theta parameters (without beta)
#' @param S0 initial population size
#'
#' @return theta including beta
#' @export
#'
add_beta <- function(theta, S0) {

      gamma <- theta[["gamma"]]
      R0    <- theta[["R0"   ]]

      theta[["beta"]] <- R0 * gamma / S0

      return(theta)

}


#' Include beta to the parameter vector theta
#'
#' @param theta parameters (without gamma)
#'
#' @return theta including beta
#' @export
#'
add_gamma <- function(theta) {

      lambda <- theta[["lambda"]]
      shape  <- theta[["shape" ]]

      theta[["gamma"]] <- 1 / (lambda ^ (- 1 / shape) * gamma(1 + 1 / shape))

      return(theta)

}


#' Include beta to the parameter vector theta
#'
#' @param theta parameters (without beta)
#' @param iota_dist distribution of infection periods
#' @param S0 initial population size
#'
#' @return theta including beta
#' @export
#'
complete_theta <- function(theta, iota_dist, S0) {

      if(iota_dist == "exponential") theta[c("shape", "lambda")] <- NULL
      if(iota_dist == "weibull")     theta <- add_gamma(theta)
      theta <- add_beta(theta, S0)

      return(theta)

}