pattRrep.fit: Function to fit a pattern model for repeated rankings)
In prefmod: Utilities to Fit Paired Comparison Models for Preferences

pattRrep.fit

R Documentation

Function to fit a pattern model for repeated rankings)

Description

Function to fit a pattern model for repeated (partial) rankings (transformed to paired comparisons) allowing for missing values using a CL approach.

Usage

pattRrep.fit(obj, nitems, tpoints = 1, formel = ~1, elim = ~1,
        resptype = "rankingT", obj.names = NULL, ia = FALSE,
        iaT = FALSE, NItest = FALSE, pr.it = FALSE)

Arguments

`obj`	either a dataframe or the path/name of the datafile to be read.
`nitems`	the number of items at one time point.
`tpoints`	the number of time points (must be > 1).
`formel`	the formula for subject covariates to fit different preference scales for the objects (see below).
`elim`	the formula for the subject covariates that specify the table to be analysed. If omitted and `formel` is not `~1` then `elim` will be set to the highest interaction between all terms contained in `formel`. If `elim` is specified, the terms must be separated by the `*` operator.
`resptype`	is `"rankingT"` by default and is reserved for future usage. Any other specification will not change the behaviour of `pattL.fit`
`obj.names`	character vector with names for objects.
`ia`	`FALSE` by default, has no meaning for rankings. Reserved for future usage.
`iaT`	if `iaT = TRUE`, dependence parameters for each item between two successive time points.
`NItest`	separate estimation of object parameters for complete and incomplete patterns if `NItest = TRUE`. Currently, `NItest` is set to `FALSE` if subject covariates are specified.
`pr.it`	a dot is printed at each iteration cycle if set to `TRUE`

Details

Models including categorical subject covariates can be fitted using the formel and elim arguments. formel specifies the actual model to be fitted. For instance, if specified as formel = ~SEX different preference scale for the objects will be estimated for males and females. For two or more covariates, the operators + or * can be used to model main or interaction effects, respectively. The operator : is not allowed (redundant terms are removed automatically). See also formula.

The specification for elim follows the same rules as for formel. However, elim specifies the basic contingency table to be set up but does not specify any covariates to be fitted. This is done using formel. If, e.g., elim = ~SEX but formel = ~1, then the table is set up as if SEX would be fitted but only one global preference scale is computed. This feature allows for the successive fitting of nested models to enable the use of deviance differences for model selection (see example below).

Value

pattRrep.fit returns an object of class pattMod. The function print (i.e., print.pattMod) can be used to print the results and the function patt.worth to produce a matrix of worth parameters.

An object of class pattMod is a list containing the following components:

`coefficients`	estimates
`ll`	log-likelihood of the model
`fl`	log-likelihood of the saturated model
`call`	function call
`result`	a list of results from the fitting routine (see Value of `nlm`).
`envList`	a list with further fit details like subject covariates design structure `covdesmat`, paired comparison response pattern matrix `Y`, etc.
`partsList`	a list of the basic data structures for each subgroup defined by crossing all covariate levels and different missing value patterns. Each element of `partsList` is again a list containing counts, missing value pattern, the CL matrix represented as the vector `s`, and the specification of the covariates. Use `str` to inspect the elements and see example below.

Input Data

The input data must have the following order (from left to right): all items at first time point, all items at second time point (with the same order as before), etc. for the other time points, optional subject covariates. The responses have to be coded as consecutive integers starting with 1 (or 0). The value of 1 (0) means highest ‘endorsement’ (agreement) according to the underlying scale. Missing values are coded as NA, rows with less than 1 valid response are removed from the fit and a message is printed.

Optional subject covariates have to be specified such that the categories are represented by consecutive integers starting with 1. Rows with missing values for subject covariates are removed from the data and a message is printed. Again, the leftmost columns in the data must be the rankings, optionally followed by columns for categorical subject covariates.

The data specified via obj are supplied using either a data frame or a datafile in which case obj is a path/filename. The input data file if specified must be a plain text file with variable names in the first row as readable via the command read.table(datafilename, header = TRUE).

Warning

The size of the table to be analysed increases dramatically with the number of items J and time points T. For rankings the number of paired comparison response categories is always two. For each time point the number of rows of the table to set up the design matrix is initially (J!. Thus, the number of rows in the design matrix is (J!) ^ T. The number of combined covariate levels and the number of missing value patterns have effects only on the run time. A (reasonable) maximum number of items for two time points is 5 or 6, for three timepoints 4, and for four to seven timepoints 3.

Note

The number of timepoints can also be regarded as different response dimensions.

Author(s)

Reinhold Hatzinger

Examples

# simulated data: 3 items, 2 timepoints
dat1 <- simR(3, 100, c(.2, .7, .1))
dat2 <- simR(3, 100, c(.5, .4, .1))
dat  <- data.frame(dat1, dat2)
res  <- pattLrep.fit(dat, nitems = 3, tpoints = 2, iaT = TRUE)
res
patt.worth(res, obj.names = LETTERS[1:3])

prefmod documentation built on June 11, 2022, 3 p.m.