bayespref (version 1.0)

pairwiseProb: Pairwise probability for differences in preference

Description

This function calculates the pairwise probability that preference among items differs. It provides the probability that row > column.

Usage

pairwiseProb(prefres = NULL, burn = 0)

Arguments

prefres

An object returned by bayesPref.

burn

A value indicating the number of burnin samples discarded.

Value

• A square matrix of pairwise comparisons giving the probability that row > column.

Details

This function examines the proportion of times a certain preference parameter is ranked over an alternate preference parameter across each post burnin MCMC step. It can be interpreted as a post-hoc test to the results obtained from bayesPref.

Examples

data(YGGV)
res <- bayesPref(pData=YGGV,mcmcL=1000)
pairwiseProb(prefres=res[[1]],burn=100) #pairwise probability for population 1