getContrasts {gnm}R Documentation

Estimated Contrasts and Standard Errors for Parameters in a gnm Model

Description

For each set in a specified list of sets of parameters from a gnm model, computes the estimated simple contrasts (i.e., differences) with the last parameter in the set, and estimated standard errors for those estimated differences.

Usage

getContrasts(model, sets = NULL, nSets = 1, ...)

Arguments

model a model object of class "gnm"
sets a vector of indices (if nSets is 1) or a list (of length nSets) of such vectors
nSets the number of vectors of indices to use
... arguments to pass to other functions

Details

The indices must all be in 1:length(coef(object)). If sets = NULL, a Tk dialog is presented for the selection of indices (model coefficients).

For each set of coefficients selected, differences with the last coefficient and their standard errors are computed. A check is performed first on the estimability of all such differences.

Value

A list (of length nSets) of 2-component lists. The first component, named summary, is a data frame containing variables estimate, se and quasi.se. The quasi.se variable is not present if either the qvcalc package is unavailable or there are fewer than two estimable differences. The second component, named relative.errors, is a character vector of length 2 containing the smallest and largest relative errors of the quasi-standard-error approximation, in the set of all simple contrasts; or NULL if quasi standard errors have not been calculated. See Firth (2003) or Firth and Menezes (2004) for details of quasi standard errors and their use.

Author(s)

David Firth

References

Firth, D (2003). Overcoming the reference category problem in the presentation of statistical models. Sociological Methodology 33, 1–18.

Firth, D and Menezes, R X de (2004). Quasi-variances. Biometrika 91, 65–80.

See Also

gnm, se, checkEstimable, qvcalc

Examples

set.seed(1)
data(yaish)

## Fit the "UNIDIFF" mobility model across education levels
unidiff <- gnm(Freq ~ educ:orig + educ:dest +
               Mult(Exp(-1 + educ), orig:dest), family = poisson,
               data = yaish)
## Examine the education multipliers (differences on the log scale):
getContrasts(unidiff, grep("Mult1.Factor1", names(coef(unidiff))))

[Package gnm version 0.7-2 Index]