summary.marg: Summary Method for Objects of Class "marg"
In marg: Approximate Marginal Inference for Regression-Scale Models

Description Usage Arguments Details Value Note See Also Examples

Returns a summary list for objects of class marg.

## S3 method for class 'marg'
summary(object, alpha = 0.05, test = NULL, all = FALSE, 
        coef = TRUE, int = ifelse((is.null(test) || all), TRUE, FALSE), 
        digits = NULL, ...)

`object`	a `marg` object. This is assumed to be the result returned by the `cond.rsm` function.
`alpha`	a vector of levels for confidence intervals; the default is 5%.
`test`	a vector of values of the parameter of interest one wants to test for. If `NULL` no test is performed. The default is `NULL`.
`all`	logical value; if `TRUE` all the information stored in the `summary.marg` object is printed, else only a subset of it. The default is `FALSE`.
`coef`	logical value; if `TRUE` the unconditional and approximate conditional/marginal MLEs are printed. The default is `TRUE`.
`int`	logical value; if `TRUE` confidence intervals are printed. The default is `TRUE`.
`digits`	the number of significant digits to be printed. The default depends on the value of `digits` set by `options`
`...`	absorbs any additional argument.

This function is a method for the generic function summary() for objects of class marg. It can be invoked by calling summary or directly summary.marg for an object of the appropriate class.

A list is returned with the following components:

`coefficients`	a 2x2 matrix containing the unconditional and approximate conditional/marginal MLEs and their standard errors.
`conf.int`	a matrix containing, for each level given in `alpha`, the upper and lower confidence bounds derived from several first and higher order test statistics. One-sided and two-sided confidence intervals are considered. See `marg.object` for details on the test statistics used.
`signif.tests`	a list with two elements. The first (`stats`) contains, for each value given in `test`, the values and tail probabilities of several first and higher order test statistics. See `marg.object` for details on the test statistics. The second element of the list (`qTerm`) contains for each tested hypothesis the correction term used in the higher order solutions.
`call`	the function call that created the `marg` object.
`formula`	the model formula.
`family`	the name of the error distribution.
`offset`	the covariate occurring in the model formula whose coefficient represents the parameter of interest or `scale` if the parameter of interest is the scale parameter.
`alpha`	the vector of levels used to compute the confidence intervals.
`hypotheses`	the values for the parameter of interest that have been tested for.
`diagnostics`	the information and nuisance parameters aspects; see `marg.object` for details.
`n.approx`	the number of output points that have been calculated exactly.
`all`	logical value; if `TRUE`, all the information stored in the `summary.marg` object is printed.
`cf`	logical value; if `TRUE`, the parameter estimates are printed.
`int`	logical value; if `TRUE`, confidence intervals are printed.
`is.scalar`	a logical value indicating whether there are any nuisance parameters. If `FALSE` there are none.
`digits`	the number of significant digits to be printed.

If the parameter of interest is the scale parameter, all calculations are performed on the log scale, though most results are reported on the original scale.

The amount of information calculated may vary depending on whether there are any nuisance parameters. A message is printed if there are none.

summary, marg.object

## House Price Data
data(houses)
houses.rsm <- rsm(price ~ ., family = student(5), data = houses)
houses.marg <- cond(houses.rsm, floor)
summary(houses.marg, test = 0, coef = FALSE)