testMod-methods: Methods for Function testMod in Package 'MAINT.Data'

In MAINT.Data: Model and Analyse Interval Data

Methods for Function testMod in Package ‘MAINT.Data’

Description

Performs statistical likelihood-ratio tests that evaluate the goodness-of-fit of a nested model against a more general one.

Usage

testMod(ModE,RestMod=ModE@ModelConfig[2]:length(ModE@ModelConfig),FullMod="Next")

Arguments

ModE	An object of class IdtE representing the estimates of a model fitted to a data set of interval-value variables
RestMod	Indices of the restricted models being evaluated in the NULL hypothesis
FullMod	Either indices of the general models being evaluated in the alternative hypothesis or the strings "Next" (default) or "All". In the former case a Restricted model is always compared against the most parsimonious alternative that encompasses it, and in latter all possible comparisons are performed

Value

An object of class ConfTests with the results of the tests performed

Examples

# Create an Interval-Data object containing the intervals of temperatures by quarter 
# for 899 Chinese meteorological stations.

ChinaT <- IData(ChinaTemp[1:8])

# Estimate by maximum likelihood the parameters of Gaussian models 
# for the Winter (1st and 4th) quarter intervals

ChinaWTE <- mle(ChinaT[,c(1,4)])
cat("China maximum likelhiood estimation results for Winter quarters:\n")
print(ChinaWTE)

# Perform Likelihood-Ratio tests comparing models with consecutive nested Configuration 
testMod(ChinaWTE)

# Perform Likelihood-Ratio tests comparing all possible models 
testMod(ChinaWTE,FullMod="All")

# Compare model with covariance Configuration case 3 (MidPoints independent of LogRanges) 
# against model with covariance Configuration 1 (unrestricted covariance)  
testMod(ChinaWTE,RestMod=3,FullMod=1)

MAINT.Data documentation built on April 4, 2023, 9:09 a.m.