bmpt.model-class: Class '"bmpt.model"'

In MPTinR2: Analyze Multinomial Processing Tree Models

Description

bmpt.model and mpt.model are two classes representing a multinomial processing tree (MPT) model. mpt.model is the superclass and bmpt.model a subclass for all those MPT models that are members of L-BMPT, the context-free language for MPT models (Purdy & Batchelder, 2009).

Objects from the Class

Objects can be created by calling make.mpt.

For a MPT to be a BMPT (or a member of L-BMPT), it is absolutely necessary that the equations representing the model perfectly map the tree structure of the (binary) MPT. That is, the model file is only allowed to contain parameters, their negations (e.g., Dn and (1 - Dn)) and the operators + and *, but nothing else. Simplifications of the equations or a change in the order of the parameters will lead to an object of class mpt.model. Models represented as mpt.models have numerous disadvantages: Fitting is a lot slower, you cannot compute the FIA, ...

To take full advantage of MPTinR specify a bmpt.model.

Slots

check:

List containing various information about the model (e.g., parameters, degrees of freedom,...).

initial.model.data.frame:

data.frame representation of the initial model (i.e., no restrictions applied). Basically a representation in EQN syntax plus a column indicating the branch number.

model.data.frame:

data.frame representation of the final model (i.e., restrictions applied). Basically a representation in EQN syntax plus a column indicating the branch number.

initial.model:

List representation of the inital model's equations (i.e., no restrictions applied).

model.list:

List representation of the final model's equations (i.e., restrictions applied).

restrictions:

List containing the restrictions (uses non-visible classes for representation of restrictions).

A:

So called A matrix. Three-dimensional array representing the position of the non-inverted parameters (e.g., Dn) in the model. First dimension = response category, second dimension = branch for that response category, third dimension = parameter (in alphabetical order). Only in bmpt.models.

B:

So called B matrix. Three-dimensional array representing the position of the inverted parameters (e.g., (1 - Dn)) in the model. First dimension = response category, second dimension = branch for that response category, third dimension = parameter (in alphabetical order). Only in bmpt.models.

lbmpt:

L-BMPT representation of the model. Only in bmpt.models.

Extends

bmpt.model extends mpt.model.

Methods

check

signature(object = "mpt.model"): Very informative function that returns a list with useful information for a model:
Do probabilities sum to 1?; number of trees; number of categories; number of free parameters in the final model; names of free parameters in the final model; number of fixed parameters in the final model; names of fixed parameters in the final model; names of parameters in the initial model (i.e., no restrictions applied); maximum number of branches per category; vector with number of branches per category; degrees of freedom; is model a member of L-BMPT?
Is part of the show method for model objects.

initial.model.data.frame

signature(object = "mpt.model"): Returns a data.frame representing the initial model (i.e., no restrictions applied).

model.data.frame

signature(object = "mpt.model"): Returns a data.frame representing the final model (i.e., restrictions applied).

lbmpt

signature(object = "bmpt.model"): returns a character vector with the L-BMPT representation.

Author(s)

Henrik Singmann

References

Purdy, B. P., & Batchelder, W. H. (2009). A context-free language for binary multinomial processing tree models. Journal of Mathematical Psychology, 53, 547-561.

See the following reference for more on A and B matrices:
Hu, X., & Batchelder, W. H. (1994). The statistical analysis of general processing tree models with the EM algorithm. Psychometrika, 59(1), 21-47.

See Also

make.mpt, fit.mpt.

Examples

showClass("bmpt.model")

MPTinR2 documentation built on May 2, 2019, 4:44 p.m.