Bootstrap for Elimination-by-Aspects (EBA) Models

Performs a bootstrap by resampling the individual data matrices.

boot(D, R = 100, A = 1:I, s = rep(1/J, J), constrained = TRUE)

Arguments

D

either a 3d array consisting of the individual paired comparison matrices or an object of class paircomp

R

the number of bootstrap samples

A

a list of vectors consisting of the stimulus aspects; the default is 1:I, where I is the number of stimuli

s

the starting vector with default 1/J for all parameters, where J is the number of parameters

constrained

logical, if TRUE (default), parameters are constrained to be positive

Details

The bootstrap functions eba.boot.constrained and eba.boot are automatically called by boot.

The code is experimental and may change in the future.

Value

p

the matrix of bootstrap vectors

stat

the matrix of bootstrap statistics, including parameter means, standard errors, and confidence limits

See also

eba, simulate.eba, paircomp.

Examples

data(pork)  # pork tasting data, 10 individual paired comparison matrices
m <- eba(apply(pork, 1:2, sum))  # fit Bradley-Terry-Luce model
b <- boot(pork, R = 200)         # resample 200 times

plot(coef(m), b$stat[, "mean"], log = "xy")
abline(0, 1, lty = 2)