stab_elo {EloRating}

stability index S

Description

calculates the S index as metric for the overall stability of a hierarchy during a specified time period

Usage

stab_elo(
  eloobject,
  from = min(eloobject$stability$date),
  to = max(eloobject$stability$date),
  weight = TRUE
)

Arguments

eloobject

an object of class "elo", usually the result of a call to elo.seq

from

character, from which date onwards should S be calculated. By default the first date in the sequence is used

to

character, until which date should S be calculated. By default the last date in the sequence is used

weight

logical, should single rank changes be weighted by the Elo rating of the highest-rated individual involved in a rank change? Default is TRUE

Details

S ranges between 0 and 1, where 0 indicates an unstable hierarchy, in which the ordering reverses every other day, and 1, in which the ordering is stable and no rank changes occur.

In contrast to the originally proposed S, this version is now standardized between 0 and 1, and additionally, the interpretation is reversed, i.e. 1 refers to stable situations, whereas values closer to 0 indicate more instable hierarchies

Value

returns the S index

Author(s)

Christof Neumann

References

Neumann C, Duboscq J, Dubuc C, Ginting A, Irwan AM, Agil M, Widdig A, Engelhardt A (2011). “Assessing dominance hierarchies: validation and advantages of progressive evaluation with elo-rating.” Animal Behaviour, 82, 911-921. doi:10.1016/j.anbehav.2011.07.016.

McDonald DB, Shizuka D (2013). “Comparative transitive and temporal orderliness in dominance networks.” Behavioral Ecology, 24, 511-520. doi:10.1093/beheco/ars192.

Examples

data(adv)
SEQ <- elo.seq(winner=adv$winner, loser=adv$loser, Date=adv$Date)
stab_elo(SEQ)
stab_elo(SEQ, weight=FALSE)
stab_elo(SEQ, from="2010-01-20", to="2010-01-30")
stab_elo(SEQ, from="2010-01-20", to="2010-01-30", weight=FALSE)