randomelo: calculate Elo ratings from an interaction matrix

In gobbios/EloRating: Animal Dominance Hierarchies by Elo Rating

calculate Elo ratings from an interaction matrix

Description

calculate Elo ratings from an interaction matrix based on randomly generated sequences

Usage

randomelo(
  interactionmatrix,
  runs = 2000,
  normprob = TRUE,
  k = 100,
  progressbar = FALSE
)

Arguments

interactionmatrix	square interaction matrix with winner in rows and losers in columns, for example the output from creatematrix
runs	number of randomly generated sequences based on the interactions in the interactionmatrix
normprob	logical (by default TRUE). Should a normal curve be assumed for calculating the winning/losing probablities, or a logistic curve. See winprob for details
k	numeric, factor k that determines the maximum change in ratings. By default k=100
progressbar	logical, should progress bars be displayed, by default progressbar=TRUE

Value

list of length 2. The first element contains a matrix with the final rating of each individual from each random sequence. IDs are in the columns, each run is represented as one row. The second element of the list contains the original interaction matrix.

Author(s)

Christof Neumann

Examples

data(adv)
elores <- elo.seq(winner = adv$winner, loser = adv$loser, Date = adv$Date)
mat <- creatematrix(elores)
res <- randomelo(mat, 10)
data.frame(ID = colnames(res[[1]]), avg = round(colMeans(res[[1]]), 1))

gobbios/EloRating documentation built on June 4, 2023, 6:33 a.m.