simulate.eba {eba} | R Documentation |
Simulates responses from the distribution corresponding to a fitted
eba
model object.
## S3 method for class 'eba'
simulate(object, nsim, seed, pool = TRUE, ...)
object |
an object of class |
nsim |
currently not used |
seed |
currently not used |
pool |
logical, if TRUE (default), pooled responses (summed across respondents) are returned |
... |
further arguments passed to or from other methods; none are used in this method |
Responses are simulated by rbinom
with sizes taken from the
original sample and probabilities computed from the model object.
A paired-comparison matrix of (pooled) responses.
data(celebrities) # absolute choice frequencies
A <- list(c(1,10), c(2,10), c(3,10),
c(4,11), c(5,11), c(6,11),
c(7,12), c(8,12), c(9,12)) # the structure of aspects
eba1 <- eba(celebrities, A) # fit elimination-by-aspects model
## Parametric bootstrap of goodness-of-fit test
LR.stat <- replicate(200, deviance(eba(simulate(eba1), A)))
hist(LR.stat, col="lightgray", border="white", freq=FALSE, breaks=20,
xlim=c(0, 60), main="Parametric bootstrap")
curve(dchisq(x, df=eba1$goodness.of.fit["df"]), add=TRUE)
abline(v=deviance(eba1), lty=2)