Simulates data based on a specified slca model. If the model parameters are not already estimated, they can either be provided by the user or generated randomly.
Usage
# S3 method for class 'slca'
simulate(object, nsim = 500, seed = NULL, parm, nlevel, ...)Arguments
- object
an
slcaobject representing the model from which data will be simulated.- nsim
a positive number specifying the number of response observations to simulate. Non-integer values are rounded up. The default is 500.
- seed
an integer specifying the random seed for reproducibility. If not provided, results will vary across runs.
- parm
a user-specified set of parameters to guide the simulation. This is required if the model has not been previously estimated. If the supplied length does not match the model, random parameters are used with a warning.
- nlevel
an integer or integer vector specifying the number of levels for each manifest item in the model. If a single integer is provided, all manifest items use that number of levels. Named vectors may specify a subset of manifest items, with unspecified items using the default of 2.
- ...
Additional arguments passed to other methods.
Value
A list with the following components:
- class
A
data.framecontaining the assigned latent class for each individual across all latent class variables.- response
A
data.framecontaining the simulated manifest item responses as factors. Internal response encodings used for model calculation are not stored on this object.
Examples
m1 <- slca(lc1[3] ~ x1 + x2 + x3 + x4 + x5,
lc2[4] ~ y1 + y2 + y3 + y4 + y5)
sim <- simulate(m1, 1000)
sapply(sim$class, table)
#> $lc1
#>
#> 1 2 3
#> 229 367 404
#>
#> $lc2
#>
#> 1 2 3 4
#> 202 137 325 336
#>
# simulate data with defined number of levels of manifest items
m2 <- slca(lc1[3] ~ x1 + x2 + x3 + x4)
sim <- simulate(m2, nlevel = c(3, 3, 3, 3))
d <- sim$response
sapply(d, table)
#> x1 x2 x3 x4
#> 1 83 211 251 156
#> 2 224 174 104 139
#> 3 193 115 145 205
sim <- simulate(m2, nlevel = c(x1 = 2, x3 = 3, x4 = 4))
d <- sim$response
sapply(d, table)
#> $x1
#>
#> 1 2
#> 276 224
#>
#> $x2
#>
#> 1 2
#> 407 93
#>
#> $x3
#>
#> 1 2 3
#> 141 143 216
#>
#> $x4
#>
#> 1 2 3 4
#> 108 180 49 163
#>
# simulate data with user-defined parameters
pi <- rep(1 / 3, 3)
rho <- c(.9, .1, .9, .1, .9, .1, .9, .1,
.9, .1, .9, .1, .1, .9, .1, .9,
.1, .9, .1, .9, .1, .9, .1, .9)
par <- c(pi, rho)
m3 <- slca(lc[3] ~ y1 + y2 + y3 + y4)
sim <- simulate(m3, parm = par)
mf <- estimate(m3, sim$response)
param(mf)
#> PI :
#> (lc)
#> class
#> 1 2 3
#> 0.3664 0.2709 0.3627
#>
#> RHO :
#> (a)
#> class
#> response 1 2 3
#> 1(V1) 0.1137 0.8846 1.0000
#> 2 0.8863 0.1154 0.0000
#> 1(V2) 0.1975 0.8930 0.8705
#> 2 0.8025 0.1070 0.1295
#> 1(V3) 0.1036 1.0000 0.2347
#> 2 0.8964 0.0000 0.7653
#> 1(V4) 0.0607 0.9880 0.1988
#> 2 0.9393 0.0120 0.8012
#>
#> V1 V2 V3 V4
#> lc y1 y2 y3 y4