broken_stick: Random generation of datasets using the dirichlet broken...

In zoid: Bayesian Zero-and-One Inflated Dirichlet Regression Modelling

Random generation of datasets using the dirichlet broken stick method

Description

Random generation of datasets using the dirichlet broken stick method

Usage

broken_stick(
  n_obs = 1000,
  n_groups = 10,
  ess_fraction = 1,
  tot_n = 100,
  p = NULL
)

Arguments

n_obs	Number of observations (rows of data matrix to simulate). Defaults to 10
n_groups	Number of categories for each observation (columns of data matrix). Defaults to 10
ess_fraction	The effective sample size fraction, defaults to 1
tot_n	The total sample size to simulate for each observation. This is approximate and the actual simulated sample size will be slightly smaller. Defaults to 100
p	The stock proportions to simulate from, as a vector. Optional, and when not included, random draws from the dirichlet are used

Value

A 2-element list, whose 1st element X_obs is the simulated dataset, and whose 2nd element is the underlying vector of proportions p used to generate the data

Examples

y <- broken_stick(n_obs = 3, n_groups = 5, tot_n = 100)

# add custom proportions
y <- broken_stick(
  n_obs = 3, n_groups = 5, tot_n = 100,
  p = c(0.1, 0.2, 0.3, 0.2, 0.2)
)

zoid documentation built on Sept. 12, 2023, 1:21 a.m.