LRT: Likelihood ratio test

In Rumenick/flexcure: Flexible Parametric Survival Models with Cure Fraction

Description

Computes the likelihood ratio test.

Usage

LRT(fits, fitg)

Arguments

fits	an object that stores the results of curereg fit of the model under the null hypothesis.
fitg	an object that stores the results of curereg fit of the model under the alternative hypothesis.

Details

The objects fitg and fits are obtained using the usual options passed to the curereg function.

Value

LRS	the value of the likelihood ratio statistic.
pvalue	the p value of test under null hypothesis chi-square distribution.

Author(s)

Rumenick Pereira da Silva [e-mail]

See Also

curereg, plot.curereg, lines.curereg

Examples

# ?rmoeevsm
df <- rmoeevsm(n = 500, mu = 0, sigma = 1, alpha = 1, theta = .2, control = 15)

# Model under the null hypothesis (sigma = 1)
fits <- curereg(Surv(time, status) ~ 1, cureformula = ~ 1,
data = df, timedist = "moee", ncausedist = "bernoulli",
method = "BFGS")

# Model under the alternative hypothesis (sigma != 1)
fitg <- curereg(Surv(time, status) ~ 1, cureformula = ~ 1,
data = df, timedist = "moeev", ncausedist = "bernoulli",
method = "BFGS")
LRT(fits, fitg)

Rumenick/flexcure documentation built on May 9, 2019, 10:37 a.m.