run_sim: Run a simulation

In haziqj/diagacc: Modelling Diagnostic Accuracy using Latent Class Models

Description

Repeatedly fit the LC, LCRE and FM model onto a specific scenaria.

Usage

run_sim(
  B = 4,
  n = 250,
  tau = 0.08,
  miss.prop = 0.2,
  data.gen = c("lc", "lcre", "fm"),
  lc.method = "MCMC",
  lcre.method = "MCMC",
  pb,
  object = NULL
)

Arguments

B	Number of replications
n	(numeric) Sample size
tau	(numeric, between 0 and 1) Prevalence
miss.prop	(numeric, between 0 and 1) Missing proportion
data.gen	The data generating mechanism
lc.method	(DEPRECATED) The latent class model estimation method
lcre.method	(DEPRECATED) The latent class model with random effects estimation method
pb	Not used
object	A diagaccSim1 object

Details

The framework for the simulation is as follows:

1. Randomly generate a specific data set according to the data.gen mechanism and the settings as per n, tau, and miss.prop.

2. Fit a LC, LCRE and FM model onto the data set.

3. Repeat steps 1–2 a total of B number of times.

Note that it is possible to continue on a simulation run by calling the saved object in the argument. See the example section.

Value

A diagaccSim1 object. It is a list of length four containing the following items:

• LC: A list of length B containing the LC model fit using the fit_lc() function.

• LCRE: A list of length B containing the LCRE model fit using the fit_lcre() function.

• FM: A list of length B containing the LC model fit using the fit_fm() function.

• sim.settings: A list of length B containing the LC model fit using the fit_lc() function.

haziqj/diagacc documentation built on Nov. 21, 2020, 5:40 p.m.