momEdge: Central moment estimates for Edgeworth expansions

In innager/edgee: Edgeworth Expansions and Higher-Order Inference

Central moment estimates for Edgeworth expansions

Description

Calculate central moment estimates for use in Edgeworth expansions for one- and two-sample t-tests.

Usage

getMomEdgeBias(smp)

getMomEdgeUnb(smp)

getMomEdgeBias2(smp, a)

getMomEdgeUnb2(smp, a)

Arguments

smp	sample.
a	vector of the same length as smp specifying categories of observations (should contain two unique values).

Details

For one-sample estimates: getMomEdgeBias() calculates regular unbiased sample variance and naive biased estimates for 3rd to 6th central moments; getMomEdgeUnb() calculates unbiased estimates of 2nd to 6th central moments. For two-sample estimates, where the two populations are assumed to have the same variance and higher central moments: getMomEdgeBias2() calculates unbiased pooled variance and naive biased 3rd to 6th central moments; getMomEdgeUnb2() provides unbiased pooled estimates of 2nd to 6th moments.

Value

A named vector of length 5. The names of the elements correspond to the estimands and are "mu2", "mu3", "mu4", "mu5", "mu6".

See Also

getLam for calculating scaled cumulants from moments.

Examples

n     <- 10 
n2smp <- 23
smp <- rgamma(n, shape = 3)        
getMomEdgeBias(smp)  # var unbiased, moments 3 - 6 naive biased
getMomEdgeUnb( smp)

smp2 <- rgamma(n2smp, shape = 3)
treatment <- sample(0:1, size = n2smp, replace = TRUE)  
getMomEdgeBias2(smp2, treatment)  # pooled var, moments 3 - 6 naive biased
getMomEdgeUnb2( smp2, treatment)
 

innager/edgee documentation built on April 24, 2024, 8:14 p.m.