simulate_random_samples: Generate a samples for a mixture.

In rubias: Bayesian Inference from the Conditional Genetic Stock Identification Model

Description

Creates random reporting unit (rho) and collection (omega) proportions, and a sim_colls vector for simulation of individual genotypes, based on the methods used in Hasselman et al. (2015)

Usage

simulate_random_samples(
  RU_starts,
  RU_vec,
  size = 100,
  alpha_repunit = 1.5,
  alpha_collection = 1.5,
  coll_sub_dirichlet_default = 1.5
)

Arguments

RU_starts	a vector delineating the reporting units in RU_vec; generated by tcf2param_list
RU_vec	a vector of collection indices, grouped by reporting unit; generated by tcf2param_list
size	The number of individuals desired in the mixture sample. Default = 100. This is ignored without a warning if alpha_repunit is specified with counts (a cnt column) or if alpha_collection is specified with counts (a cnt column).
alpha_repunit	If a vector, this is the dirichlet parameter for simulating the proportions of reporting units. Gets recycled to the number of reporting units. Default is 1.5. Otherwise, this could be a two-column data frame. The first column must be named "repunit" and the second one must be one of "dirichlet", "ppn", or "cnt", according to whether you wish to specify dirichlet parameters, or proportions, or exact counts, respectively, for each population.
alpha_collection	The dirichlet parameter for simulating proportions of collections within reporting units. Default = 1.5
coll_sub_dirichlet_default	If you are providing a data frame with requested sub_dirichlet pars for collections, and you don't specifically list one, this is the value it gets.

Value

a list with three elements. The first two are a rho vector and an omega vector, respectively. The third is a vector of origins for simulated individuals, sampled from the collections with probabilities = omega

rubias documentation built on Feb. 10, 2022, 1:06 a.m.