psychotools (version 0.4-0)

mptmodel: Multinomial Processing Tree (MPT) Model Fitting Function

Description

mptmodel is a basic fitting function for multinomial processing tree (MPT) models.

Usage

mptmodel(y, weights = NULL, spec, treeid = NULL, optimargs = list(control = list(reltol = .Machine$double.eps^(1/1.2), maxit = 1000)), start = NULL, vcov = TRUE, estfun = FALSE, ...)

Arguments

y

matrix of response frequencies.

weights

an optional vector of weights (interpreted as case weights).

spec

an object of class mptspec: typically result of a call to mptspec. A symbolic description of the model to be fitted.

treeid

a vector that identifies each tree in a joint multinomial model.

optimargs

a list of arguments passed to the optimization function (optim).

start

a vector of starting values for the parameter estimates between zero and one.

vcov

logical. Should the estimated variance-covariance be included in the fitted model object?

estfun

logical. Should the empirical estimating functions (score/gradient contributions) be included in the fitted model object?

...

further arguments passed to functions.

Value

mptmodel returns an S3 object of class "mptmodel", i.e., a list with components as follows:

y

a matrix with the response frequencies,

coefficients

estimated parameters,

loglik

log-likelihood of the fitted model,

npar

number of estimated parameters,

weights

the weights used (if any),

nobs

number of observations (with non-zero weights),

ysum

the aggregate response frequencies,

fitted, goodness.of.fit, ...

see mpt in the mpt package.

Details

mptmodel provides a basic fitting function for multinomial processing tree (MPT) models, intended as a building block for fitting MPT trees in the psychotree package. While mptmodel is intended for individual response frequencies, the mpt package provides functions for aggregate data.

MPT models are specified using the mptspec function. See the documentation in the mpt package for details. mptmodel returns an object of class "mptmodel" for which several basic methods are available, including print, plot, summary, coef, vcov, logLik, estfun and predict.

See Also

btmodel, pcmodel, rsmodel, raschmodel, mptspec, the mpt package

Examples

o <- options(digits = 4)

## data
data("SourceMonitoring", package = "psychotools")

## source-monitoring MPT model
mpt1 <- mptmodel(SourceMonitoring$y, spec = mptspec("SourceMon"))
summary(mpt1)
plot(mpt1)

options(digits = o$digits)