FitStack: Up-dates association parameters of the nominal model

Description

Discrete choice model (conditional multinominal logistic regression model) is fit to stacked data to up-date the matrix of association parameters of the LMA that corresponds to the nominal item response model. This is a function internal to 'fit.nominal' and is used for multi-dimensional models. The function is similar to 'fit.StackGPCM'. This function is unlikely to be run outside of 'fit.nominal' or 'ple.lma'.

Usage

FitStack(
  Master,
  item.log,
  phi.log,
  fstack,
  TraitByTrait,
  pq.mat,
  npersons,
  nitems,
  ncat,
  nless,
  ntraits,
  Maxnphi,
  PhiNames,
  LambdaNames
)

Arguments

Master

Master data set from which stacked data is created

item.log

Last row contains current scale values (item.history)

phi.log

Last row contains current estimates of phi

fstack

Formula for stacked regression

TraitByTrait

inTraitAdj matrix

pq.mat

Summing array to get rest scores and totals

npersons

Number of persons

nitems

Number of items

ncat

Number of categories per item

nless

Number of categories less 1 (unique lambdas & unique nus)

ntraits

Number of latent traits

Maxnphi

Number of phis to be estimated

PhiNames

Names of the Phi parameters

LambdaNames

Names of lambdas that correspond to those in Master

Value

Phi.mat Matrix of up-dated estimates of assocation (phi) parameters

phi.log History of iterations log likelihood and estimates of lambda and phi parameters