Standard Errors of Linear Parameter Combinations in gnm Models

Share:

Description

Computes approximate standard errors for (a selection of) individual parameters or one or more linear combinations of the parameters in a gnm (generalized nonlinear model) object. By default, a check is made first on the estimability of each specified combination.

Usage

1
2
se(model, estimate = NULL, checkEstimability = TRUE, Vcov =
NULL, dispersion = NULL, ...)

Arguments

model

a model object of class "gnm".

estimate

(optional) specifies parameters or linear combinations of parameters for which to find standard errors. In the first case either a character vector of names, a numeric vector of indices or "[?]" to select from a Tk dialog. In the second case coefficients given as a vector or the rows of a matrix, such that NROW(estimate) is equal to length(coef(model)). If missing, standard errors are returned for all (non-eliminated) parameters in the model.

checkEstimability

logical: should the estimability of all specified combinations be checked?

Vcov

either NULL, or a matrix

dispersion

either NULL, or a positive number

...

possible further arguments for checkEstimable.

Value

A data frame with two columns:

Estimate

The estimated parameter combinations

Std. Error

Their estimated standard errors

If available, the column names of coefMatrix will be used to name the rows.

Note

In the case where estimate is a numeric vector, se will assume that indices have been specified if all the values of estimate are in seq(length(coef(model)).

Where both Vcov and dispersion are supplied, the variance-covariance matrix of estimated model coefficients is taken to be Vcov * dispersion.

Author(s)

David Firth and Heather Turner

See Also

gnm, getContrasts, checkEstimable, ofInterest

Examples

 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14
15
16
17
set.seed(1)

## Fit the "UNIDIFF" mobility model across education levels
unidiff <- gnm(Freq ~ educ*orig + educ*dest +
               Mult(Exp(educ), orig:dest),
               ofInterest = "[.]educ", family = poisson,
               data = yaish,  subset = (dest != 7))
## Deviance is 200.3

## Get estimate and se for the contrast between educ4 and educ5 in the
## UNIDIFF multiplier
mycontrast <- numeric(length(coef(unidiff)))
mycontrast[ofInterest(unidiff)[4:5]] <- c(1, -1)
se(unidiff, mycontrast)

## Get all of the contrasts with educ5 in the UNIDIFF multipliers
getContrasts(unidiff, rev(ofInterest(unidiff)))

Want to suggest features or report bugs for rdrr.io? Use the GitHub issue tracker.