selm-class {sn}

Classes selm and mselm of objects created by function selm

Description

A successful call to function selm creates an object of either of these classes, having a structure described in section ‘Slots’. A set of methods for these classes of objects exist, listed in section ‘Methods’.

Objects from the class

An object can be created by a successful call to function selm.

Slots

call:

the calling statement.

family:

the parametric family of skew-ellitically contoured distributed (SEC) type.

logL:

log-likelihood or penalized log-likelihood value achieved at the end of the maximization process.

method:

estimation method ("MLE" or "MPLE").

param:

estimated parameters, for various parameterizations.

param.var:

approximate variance matrices of the parameter estimates, for various parameterizations.

size:

a numeric vector with size of various components.

fixed.param:

a vector of parameters which have been kept fixed in the fitting process, if any.

residuals.dp:

residual values, for DP-type parameters.

fitted.values.dp:

fitted values, for DP-type parameters.

control:

a list with control parameters.

input:

a list of selected input values.

opt.method:

a list with details on the optimization method.

Methods

coef	signature(object = "selm"): ...
logLik	signature(object = "selm"): ...
plot	signature(x = "selm"): ...
show	signature(object = "selm"): ...
summary	signature(object = "selm"): ...
residuals	signature(object = "selm"): ...
fitted	signature(object = "selm"): ...
vcov	signature(object = "selm"): ...
weights	signature(object = "selm"): ...
profile	signature(fitted = "selm"): ...
confint	signature(object = "selm"): ...
predict	signature(object = "selm"): ...
coef	signature(object = "mselm"): ...
logLik	signature(object = "mselm"): ...
plot	signature(x = "mselm"): ...
show	signature(object = "mselm"): ...
summary	signature(object = "mselm"): ...
residuals	signature(object = "mselm"): ...
fitted	signature(object = "mselm"): ...
vcov	signature(object = "mselm"): ...
weights	signature(object = "mselm"): ...

Note

See dp2cp for a description of possible parameter sets. When logLik is used on an object obtained using the MPLE estimation method, the value reported is actually the penalized log-likelihood.

Author(s)

Adelchi Azzalini

See Also

See also selm function, plot.selm, summary.selm, dp2cp

Examples

data(ais)
m1 <- selm(log(Fe) ~ BMI + LBM, family="SN", data=ais)
summary(m1)
plot(m1)
logLik(m1)
res <- residuals(m1)
fv <- fitted(m1)
# 
data(wines, package="sn")
m2 <- selm(alcohol ~ malic + phenols, data=wines)
#
m12 <- selm(cbind(acidity, alcohol) ~ phenols + wine, family="SN", data=wines)
coef(m12)
cp <- coef(m12, vector=FALSE)
dp <- coef(m12, "DP", vector=FALSE)
plot(m12)
plot(m12, which=2, col="gray60", pch=20)