mptEM: EM Algorithm for Multinomial Processing Tree Models

In mpt: Multinomial Processing Tree Models

EM Algorithm for Multinomial Processing Tree Models

Description

Applies the EM algorithm to fit a multinomial processing tree model.

Usage

mptEM(theta, data, a, b, c, maxit = 1000, tolerance = 1e-8, 
      stepsize = 1, verbose = FALSE)

Arguments

theta	a vector of starting values for the parameter estimates.
data	a vector of absolute response frequencies.
a	a three-dimensional array representing the model structure.
b	a three-dimensional array representing the model structure.
c	a matrix of structural constants.
maxit	the maximum number of iterations.
tolerance	the convergence criterion; the iterations converge when logLik - logLik.old < tolerance.
stepsize	the step size defaulting to 1; slightly larger values may speed up convergence, but may also give errors; use with care.
verbose	logical indicating if output should be produced for each iteration.

Details

Usually, mptEM is automatically called by mpt.

A prerequisite for the application of the EM algorithm is that the probabilities of the i-th branch leading to the j-th category take the form

p_{ij}(\Theta) = c_{ij} \prod_{s = 1}^S \vartheta_s^{a_{ijs}} (1 - \vartheta_s)^{b_{ijs}},

where \Theta = (\vartheta_s) is the parameter vector, a_{ijs} and b_{ijs} count the occurrences of \vartheta_s and 1 - \vartheta_s in a branch, respectively, and c_{kj} is a nonnegative real number. The branch probabilities sum up to the total probability of a given category, p_j = p_{1j} + \dots + p_{Ij}. This is the structural restriction of the class of MPT models that can be represented by binary trees. Other model types have to be suitably reparameterized for the algorithm to apply.

See Hu and Batchelder (1994) and Hu (1999) for details on the algorithm.

Value

theta	the vector of parameter estimates.
loglik	the log-likelihood at termination of the algorithm.
pcat	a vector of predicted probabilities for each response category.
pbranch	a vector of predicted branch probabilities.
iter	the number of iterations of the algorithm.

References

Hu, X. (1999). Multinomial processing tree models: An implementation. Behavior Research Methods, Instruments, & Computers, 31(4), 689–695. \Sexpr[results=rd]{tools:::Rd_expr_doi("10.3758/BF03200747")}

Hu, X., & Batchelder, W.H. (1994). The statistical analysis of general processing tree models with the EM algorithm. Psychometrika, 59(1), 21–47. \Sexpr[results=rd]{tools:::Rd_expr_doi("10.1007/bf02294263")}

Riefer, D.M., Knapp, B.R., Batchelder, W.H., Bamber, D., & Manifold, V. (2002). Cognitive psychometrics: Assessing storage and retrieval deficits in special populations with multinomial processing tree models. Psychological Assessment, 14(2), 184–201. \Sexpr[results=rd]{tools:::Rd_expr_doi("10.1037/1040-3590.14.2.184")}

See Also

mpt.

Examples

# Fit storage-retrieval pair-clustering model to data in Riefer et al.
# (2002) using EM algorithm
mpt(mptspec("SR2"), c(243, 64, 58, 55), method = "EM")

mpt documentation built on Oct. 11, 2024, 5:07 p.m.