gammaSim {hbmem}

Function gammaSim

Description

Simulates data from a hierarchical Gamma model.

Usage

gammaSim(NN=1,NS=2,I=30,J=200,K=6,muN=log(.65),s2aN=.2,s2bN=.2,
muS=log(c(.8,1.2)),s2aS=.2,s2bS=.2,lagEffect=-.001,shape=2,
crit=matrix(rep(c(.3,.6,1,1.2,1.6),each=I),ncol=(K-1)))

Arguments

NN

Number of conditions for new words.

NS

Number of conditions for studied words.

I

Number of participants.

J

Number of items.

K

Number of response options.

muN

Mean of new-item distribution. If NN is greater than 1, then muN must be a vector of length NN.

s2aN

Variance of participant effects on mean of new-item distribution.

s2bN

Variance of item effects on mean of new-item distribution.

muS

Mean of studied-item distribution. If NS is greater than 1, then muS must be a vector of length NS.

s2aS

Variance of participant effects on mean of studied-item distribution.

s2bS

Variance of item effects on mean of studied-item distribution.

lagEffect

Linear slope of lag effect on log of studied-item scale.

shape

Common shape for both new and studied distributuions.

crit

Matrix of criteria (not including -Inf or Inf). Columns correspond to criteria, rows correspond to participants.

Value

The function returns an internally defined "uvsdSim" structure.

Author(s)

Michael S. Pratte

References

See Pratte, Rouder, & Morey (2009)

See Also

hbmem

Examples

library(hbmem)
#Data from hiererchial model
sim=gammaSim() 
slotNames(sim) 
table(sim@resp,sim@cond,sim@Scond)

#Usefull to make data.frame for passing to model-fitting functions
dat=as.data.frame(cbind(sim@subj,sim@item,sim@cond,sim@Scond,sim@lag,sim@resp))
colnames(dat)=c("sub","item","cond","Scond","lag","resp")

table(dat$resp,dat$cond,dat$Scond)