Computes a Likelihood Ratio Test

DESCRIPTION:

Computes a likelihood ratio test in a maximum likelihood problem.

This function is generic (see Methods); method functions can be written to handle specific classes of data. Classes which already have methods for this function include: slm

USAGE:

lrt(object, ...) 

REQUIRED ARGUMENTS:

object
any object, usually a fitted model object of some kind.

OPTIONAL ARGUMENTS:

...
some methods have additional arguments.

VALUE:

the return value can vary with the generic routine, but in general a list containing the computed (asymptotic) chi-squared statistic, its p-value, and the degrees of freedom is returned.

DETAILS:

A likelihood ratio test compares two or more likelihoods to examine whether the null hypothesis or an alternative hypothesis provides a significantly better fit to the observed data. Discussion of such tests can be found in most elementary statistics texts.

SEE ALSO:

, .

EXAMPLES:

object <- slm(sid.ft ~ nwbirths.ft, cov.family=CAR, data=sids, subset= -4, 
              spatial.arglist=list(weights=1/sids$births,  
              neighbor=sids.neighbor)) 
lrt(object, 0.12, c(1.3333, NA))