dglars (version 2.1.7)

logLik.dglars: Extract Log-Likelihood

Description

logLik method for an object with class ‘dglars’.

Usage

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

Value

logLik.dglars returns an object of class “loglik_dglars”. This is a named list containing the following components:

loglik

the log-likelihood of the sequence of models fitted by dglars method.

df

the numbers of non-zero estimates corresponding to the used \(\gamma\)-values.

object

the fitted dglars object.

g

the sequence of \(\gamma\)-values.

phi

a description of the estimator used to estimate the dispersion pamater.

phih

the sequence of estimated dispersion parameter.

Arguments

object

any fitted dglars object from which the log-likelihood values can be extracted.

phi

a description of the estimator used to estimate the dispersion pamater (see below for more details).

g

vector of values of the tuning parameter.

...

further arguments passed to the function grcv.

Author

Luigi Augugliaro
Maintainer: Luigi Augugliaro

Details

logLik.dglars returns the sequence of the log-likelihood values of the models fitted by “dglars”. For the binomial and Poisson family, the dispersion parameter is assumed known and equal to one while for the other families the dispersion parameter is estimated using the method specified by the argument “phi” (see phihat for more details). The optional argument g is used to specified the values of the tuning parameter; if not specified (default), the log-likelihood values are computed for the sequence of models storage in object otherwise predict.dglars is used to compute the estimate of the parameters needed to evaluate the log-likelihood function (see the example below).

See Also

dglars, phihat, predict.dglars, AIC.dglars and BIC.dglars.

Examples

#################################
# y ~ Poisson

library(dglars)
set.seed(123)
n <- 100
p <- 5
X <- matrix(abs(rnorm(n*p)),n,p)
eta <- 1 + X[, 1]  + X[, 2]
mu <- poisson()$linkinv(eta)
y <- rpois(n, mu)
out <- dglars(y ~ X, poisson)
logLik(out)
logLik(out, g = seq(2, 0.5, by = -0.1))

############################
# y ~ Gamma

n <- 100
p <- 5
X <- matrix(abs(rnorm(n*p)),n,p)
eta <- 1 + 2 * X[, 1L]
mu <- drop(Gamma()$linkinv(eta))
shape <- 0.5
phi <- 1 / shape
y <- rgamma(n, scale = mu / shape, shape = shape)
out <- dglars(y ~ X, Gamma("log"))

# generalized Pearson statistic
logLik(out, phi = "pearson")

# deviance estimator
logLik(out, phi = "deviance")

# mle estimator
logLik(out, phi = "mle")

# grcv estimator
logLik(out, phi = "grcv")