Create Observation Function Configuration — make_obs

Constructs a configuration list specifying how to compare observed and modeled data. This configuration is passed to the observation likelihood generator function (generate_incidence_comparison()) and determines which outcomes are compared and how.

Usage

make_obs_config(
  use_monthly = TRUE,
  age_group = "u5",
  include_prev = FALSE,
  use_SMC_as_covariate = FALSE,
  log_link = FALSE,
  include_pop_growth = FALSE
)

Arguments

use_monthly: Logical. If TRUE, observation and model outputs are matched at monthly resolution; otherwise, weekly data is assumed.
age_group: Character. One of "u5", "o5", "sep_ages", "total", or "all_ages". This determines which age group(s) are included in the likelihood.
include_prev: Logical. If TRUE, the observation function includes a prevalence likelihood term.
use_SMC_as_covariate: Logical. If TRUE, the incidence is adjusted using a regression effect (beta_1) on SMC coverage.
log_link: Logical. If TRUE, a log-link is used to transform the predicted incidence when adjusting by SMC coverage. If FALSE, a linear form is used.
include_pop_growth: Logical. If TRUE, model-predicted incidence is rescaled to account for population growth using the r_C and/or r_A outputs from the model.

Value

A named list containing configuration options to pass to the observation likelihood function.

Examples

obs_config <- make_obs_config(
  use_monthly = TRUE,
  age_group = "u5",
  include_prev = FALSE,
  use_SMC_as_covariate = TRUE,
  log_link = TRUE,
  include_pop_growth = TRUE
)