sim.data: Create data based on Thurstonian model equations

Description

Simulates a data set of paired comparisons or ranking data based a Thurstonian latent utility model.

Usage

sim.data(nfactor = 1, nitem, nperson, itf, model = "factor",
  variables = NULL, ints = NULL, lmu = NULL, ivarcov = NULL,
  loadings = NULL, varcov = NULL, graded = FALSE, ncat = NULL,
  thres = NULL, transitive = TRUE, var = 0, fvalues = FALSE,
  sim = TRUE)

Value

Returns a list containing the true factor scores and the data, or a matrix containing the data.

Arguments

nfactor: The number of factors. Defaults to 1.
nitem: The number of items.
nperson: The number of data points (= respondents) to simulate.
itf: A vector defining the items-to-factor relation. For example `c(1,1,1,2,2,2)` defines six items, the first three correspond to factor 1, the second three correspond to factor 2.
model: The model class from which to simulate. Options are 'simple', 'factor', 'uc' and 'irt'. Defaults to 'factor'.
variables: A vector containing the names of paired comparison variables to return. If `NULL` (default), all variables are returned.
ints: A vector defining the latent intercepts of item pairs.
lmu: A vector defining the latent means of items.
ivarcov: A matrix defining the variance-covariance matrix of the items.
loadings: A vector defining the loadings of items.
varcov: A matrix defining the variance-covariance matrix of the traits.
graded: Logical. Should a graded preference model be simulated? Defaults to `FALSE`.
ncat: Numerical. The number of categories to specify for graded preference models. If graded = `FALSE`, ncat is 2.
thres: A vector of thresholds to categorize the latent difference response. If no thresholds are given, they are simulated from the distribution of the latent items. If graded = `FALSE`, the thresholds are all 0.
transitive: Logical. Should the data be transitive? If `TRUE`, ranking data is simulated, else paired comparison data is simulated. Defaults to `TRUE`.
var: A vector containing the variances for each paired comparison. Defaults to 0.
fvalues: Logical. Should simulated factor values be returned? Defaults to `FALSE`.
sim: Logical. Should the simulated data be returned? Defaults to `TRUE`.

Details

The syntax currently is able to simulate data from the latent utility model (`'simple'` and `'factor'`; Maydeu-Olivares & Böckenholt, 2005) the unconstrained factor model (`'uc'`; Maydeu-Olivares & Böckenholt, 2005) and the IRT model(`'irt'`; Maydeu-Olivares & Brown, 2010).

Examples

Run this code


nfactor <- 3
nitem <- 15
nperson <- 1000
itf <- rep(1:3,5)
varcov <- diag(1,3)

# latent utility means
set.seed(69)
lmu <- runif(nitem, -1, 1)
loadings <- runif(nitem, 0.30, 0.95)

FC <- sim.data(nfactor=nfactor, nitem=nitem, nperson=nperson, itf=itf, 
varcov = varcov, lmu = lmu, loadings=loadings)

Run the code above in your browser using DataLab