wilks.test: Estimator test based on Wilks' theorem

In WLinfer: Statistical Inference in Weighted Lindley Distribution

Estimator test based on Wilks' theorem

Description

This is a test based on Wilks' theorem, to determine which parameter space the estimated parameter is included in.

Usage

wilks.test(x, estimator, side = "two")

Arguments

x	a numeric vector or data frame.
estimator	a numeric vector with estimated lambda and phi.
side	a character string which selects the direction of wilks' theorem test ("two", "less" or "greater").

Details

By Wilks' theorem, we can test the k-dimensional parameter with chi-square distribution. The Wilks' theorem test can be performed by assuming the parameter space of null hypothesis and setting the part not included in it as the parameter space of the alternative hypothesis.

Value

wilks.test returns a list with these components:

side	a character string of one of "two", "less" or "greater".
stat	a numeric value the statistics from Wilks' theorem.
pvalue	a numeric value the p-value of the statistics.

Examples

data <- fail_fiber
mme <- MME_WL(data)
wilks.test(data, mme, side="two")
wilks.test(data, mme, side="less")
wilks.test(data, mme, side="greater")

WLinfer documentation built on Sept. 2, 2022, 9:06 a.m.