BradleyTerry: Fit a Bradley-Terry model

Description Usage Arguments Details Value References See Also Examples

View source: R/BradleyTerry.R

Description

Fits Bradley–Terry model, also known as a quasi-symmetry model, to paired comparison data via maximum (quasi) likelihood estimation.

Usage

1

Arguments

C

a square matrix of paired comparisons

family

a description of the error distribution and link function to be used. See family

Details

This implementation avoids calling BradleyTerry2::BTm() and instead fits the glm directly. By default, the model is fitted by quasi-likelihood (allowing non-integer counts) with a logit link. To fit by maximum likelihood instead, set the family argument to binomial.

Value

An object of class "glm"

References

Bradley, R. A., & Terry, M. E. (1952). Rank analysis of incomplete block designs: I. The method of paired comparisons. Biometrika, 39(3/4), 324–345.

See Also

Other network centrality estimators: BTscores, ILSR, PageRank, Scroogefactor

Other Bradley-Terry model utility functions: hessianBT, loglikelihood, pseudocount, vcovBT

Examples

1

Selbosh/scrooge documentation built on May 5, 2019, 8 p.m.