likelihoodRatioTest: likelihoodRatioTest

In mrparker909/mvregress: Classic Multivariate Linear Regression

Description

Calculates the likelihood ratio test statistic and associated p-value under the chi squared approximation with degrees of freedom df.

Usage

likelihoodRatioTest(logLik_H0, logLik_H1, df)

Arguments

logLik_H0	log likelihood of the null model
logLik_H1	log likelihood of the alternative model
df	degrees of freedom for the chisquared random variable (usually the difference in degrees of freedom between the alternative and the null models)

Examples

y1 = seq(0.1,1,0.1)
y2 = y1*y1+1

Y = cbind(y1,y2)

x1 = log(1:10)
x2 = x1^2-x1
X = cbind(x1,x2)

LL_H1 = logLikMVR(lm(Y~X)) # here degrees of freedom is 3*2 = 6
LL_H0 = logLikMVR(lm(Y~1)) # here degrees of freedom is 1*2 = 2
likelihoodRatioTest(LL_H0, LL_H1, df=6-2)

mrparker909/mvregress documentation built on Dec. 27, 2019, 3:40 p.m.