dglars (version 2.1.7)

coef.dglars: Extract the dgLARS Coefficient Path

Description

coef.dglars is used to extract the coefficient path computed by dgLARS method.

Usage

# S3 method for dglars
coef(object, type = c("pearson", "deviance", "mle", "grcv"),
g = NULL, ...)

Value

coef.dglars returns a named list with component:

beta

the sequence of the penalized estimates of the regression coefficients;

phi

the penalized estimates of the dispersion parameter;

g

the vector of the values of the tuning parameter.

Arguments

object

fitted dglars object.

type

a description of the estimator used for the dispersion parameter.

g

vector of values of the tuning parameter.

...

further arguments passed to the function link{grcv}.

Author

Luigi Augugliaro
Maintainer: Luigi Augugliaro

Details

coef.dglars is a wrapper function calling “predict.dglars” and “phihat”. By default, this function returns the sequence of the penalized coefficients and the sequence of the penalized estimate of the dispersion parameter \(\phi\). The user can specify the argumnets of the function grcv by the argument ...).

See Also

predict.dglars, phihat and grcv.

Examples

###########################
# Logistic regression model
set.seed(123)
n <- 100
p <- 10
X <- matrix(rnorm(n * p), n, p)
b <- 1:2
eta <- b[1] + X[, 1] * b[2]
mu <- binomial()$linkinv(eta)
y <- rbinom(n, 1, mu)
fit <- dglars(y ~ X, family = binomial)
coef(fit)
coef(fit, g = seq(4, 0.5, length = 10))

###########################
# Gamma family
n <- 100
p <- 10
X <- matrix(abs(rnorm(n * p)), n, p)
b <- 1:2
eta <- b[1] + X[, 1] * b[2]
mu <- drop(Gamma()$linkinv(eta))
shape <- 0.5
phi <- 1 / shape
y <- rgamma(n, scale = mu / shape, shape = shape)
fit <- dglars(y ~ X, Gamma("log"))
coef(fit, type = "pearson")
coef(fit, type = "deviance")
coef(fit, type = "mle")