gwer (version 1.0)

elliptical.diag: Diagnostic for Elliptical Regression Models

Description

This function obtains the values of the residuals and calculates the diagnostic measures for elliptical regression models.

Usage

elliptical.diag(ellipticalfit, weighting = "observed", ...)

Arguments

ellipticalfit

fit object for elliptical regression model.

weighting

type of model weighting used.

...

arguments to be used to form the default control argument if it is not supplied directly.

Value

returns a list of diagnostic arrays:

resid

ordinal residuals for the fit model.

rs

studentized residuals for the fit model.

dispersion

coefficient of dispersion for the model fit.

GL

generalized leverage for the model fit.

GLbeta

generalized leverage of location parameters estimation for the model fit.

GLphi

generalized leverage of dispersion parameters estimation for the model fit.

Bi

generalized leverage weighted by dispersion for the model fit.

Om

observed fisher information matrix of the model fit.

Iom

expected fisher information matrix of the model fit.

a, b, c

the value of D(a), D(b) and D(c), respectively, for the model fit.

Cmax

matrix of local influence for additive perturbation in response.

Lmax

matrix of local influence on coefficients (additive perturbation in predictors).

Cic

matrix of local influence for case-weight perturbation (Ci).

dmax

matrix of local influence for case-weight perturbation (dmax).

dmaxc

matrix of local influence for case-weight perturbation (|dmax|).

Ci

matrix of local influence on the scale.

Cih

main diagonal of the matrix of local influence on the scale.

h

main diagonal of the hat matrix.

References

Galea, M., Paula, G. A., and Cysneiros, F. J. A. (2005). On diagnostics in symmetrical nonlinear models. Statistics & Probability Letters, 73(4), 459-467. https://doi.org/10.1016/j.spl.2005.04.033

See Also

elliptical

Examples

# NOT RUN {
data(luzdat)
y <- luzdat$y
x1 <- luzdat$x1 ; x1 <- factor(x1) ; x1 <- C(x1,treatment)
x2 <- luzdat$x2
x3 <- (luzdat$x2)^2
luz <- data.frame(y,x1,x2,x3)
elliptical.fitt <- elliptical(y ~ x1+x2+x3, family = Student(df=5),
data=luz)
elliptical.diag(elliptical.fitt)
# }