set.up {pleLMA} | R Documentation |
Sets up the data based on input data and model specifications
Description
This function sets up the data and sets constants that are essentially the same for all models. This is used within the main wrapper function ‘ple.lma’, but can also be run independently. If a user wants to run the functions ‘fit.independence’, ‘fit.rasch’, ‘fit.gpcm’, or ‘fit.nominal’, the set up function should be run prior to using these functions to create required input. Such an approach can speed up replication studies because ‘set.up’ would only need to be run once and the response vector (i.e., named ‘y’) in the Master data frame be replaced by a new one.
Usage
set.up(
inData,
model.type,
inTraitAdj = NULL,
inItemTraitAdj = NULL,
tol = NULL,
starting.sv = NULL,
starting.phi = NULL
)
Arguments
inData |
A person x item Data frame with response patterns |
model.type |
Type of model to be fit |
inTraitAdj |
Trait x Trait adjacency matrix (NULL for independence) |
inItemTraitAdj |
Item x Trait adjacency matrix (NULL for independence) |
tol |
Tolerence for deteriming convergence (default: 1e-06) |
starting.sv |
Starting category scale values/fixed scores (default: sum equal to zero and sum of squares equal to 1) |
starting.phi |
optional: Starting phi matrix (default: identity matrix) |
Value
PersonByItem inData (rows are response patterns)
TraitByTrait Trait x Trait adjacency matrix
ItemByTrait Item x Trait adjacency matrix
item.by.trait Need for re-scaling phi.mat
starting.sv An item by number of category matrix with starting values for scale values for nominal model and fixed category scores for gpcm and rasch models
ItemNames Names of items in inData and PersonByItem
LambdaName Short list of lambda names needed for item regressions
NuName Short list of nu names names needed for item regressions
LambdaNames Long list of lambdas using in Master data set
NuNames Long list of nu using in Master data set
PhiNames Names of the unique phi parameters
npersons Number of individual or persons in data
nitems Number of items
ncat Number of categories
nless Number of unique lambdas and unique nus
ntraits Number of traits
Maxnphi Number of phis to estimate
Nstack Length of master data set
pq.mat An array used to computed (weighted) rest-scores
Phi.mat A number of traits x number of traits Phi matrix (defual: the identity matrix)
Master Master data set formated for input to to mlogit
tol Tolerence for deteriming convergence
Examples
data(dass)
inData <- dass[1:250,c("d1", "d2", "d3", "a1","a2","a3","s1","s2","s3")]
#--- to set data up for model of independence
ind.setup <- set.up(inData, model.type="independence")
#--- for model specification for uni-dimensional models
inTraitAdj <- matrix(1, nrow=1, ncol=1)
inItemTraitAdj <- matrix(1, nrow=9, ncol=1)
i.setup <- set.up(inData, model.type='independence')
r.setup <- set.up(inData, model.type='rasch', inTraitAdj,
inItemTraitAdj)
g.setup <- set.up(inData, model.type='gpcm', inTraitAdj,
inItemTraitAdj)
n.setup <- set.up(inData, model.type='nominal', inTraitAdj,
inItemTraitAdj)