summary.rsm {marg}

Summary Method for Regression-Scale Models

Description

Returns a summary list for a fitted regression-scale model.

Usage

## S3 method for class 'rsm'
summary(object, correlation = FALSE, digits = NULL, ...)

Arguments

object

a fitted rsm object. This is assumed to be the result of some fit that produces an object inheriting from the class rsm, in the sense that the components returned by the rsm function will be available.

correlation

logical argument. If TRUE, the correlation matrix for the coefficients is computed; default is TRUE.

digits

a non-null value specifies the minimum number of significant digits to be printed in values. If NULL, the value of digits set by options is used.

...

absorbs any additional argument.

Details

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

Value

A list is returned with the following components:

coefficients

a matrix with four columns, containing the coefficients, their standard errors, the z-values (or Wald statistics) and the associated p-values based on the standard normal approximation to the distribution of the z statistic.

dispersion

the value of the scale parameter used in the computations.

fixed

a logical value. If TRUE, the scale parameter is known.

residuals

the response residuals.

cov.unscaled

the unscaled covariance matrix, i.e, a matrix such that multiplying it by the squared scale parameter, or an estimate thereof, produces an estimated (asymptotic) covariance matrix for the coefficients.

correlation

the computed correlation matrix for the coefficients in the model.

family

the entire family.rsm object used.

loglik

the computed log likelihood.

terms

an object of mode expression and class term summarizing the formula.

df

the number of degrees of freedom for the model and for the residuals.

iter

the number of IRLS iterations used to compute the estimates.

nas

a logical vector indicating which, if any, coefficients are missing.

call

an image of the call that produced the rsm object, but with the arguments all named and with the actual formula.

digits

the value of the digits argument.

See Also

rsm.object, summary, rsm

Examples

## House Price Data
data(houses)
houses.rsm <- rsm(price ~ ., family = student(5), data = houses)
summary(houses.rsm)